The Pennsylvania State University
The Graduate School
College of Engineering
SELF-ASSEMBLY AND CHEMICAL PATTERNING ON GERMANIUM(100)
A Thesis in
Engineering Science
by
Jeffrey A. Lawrence
2010 Jeffrey A. Lawrence
Submitted in Partial Fulfillment
of the Requirements
for the Degree of
Master of Science
August 2010
The thesis of Jeffrey A. Lawrence was reviewed and approved* by the following:
Paul S. Weiss
Distinguished Professor of Chemistry and Physics
Thesis Co-Advisor
Mark W. Horn
Associate Professor of Engineering Science and Mechanics
Thesis Co-Advisor
Tony J. Huang
Associate Professor of Engineering Science and Mechanics
Judith A. Todd
P.B. Breneman Department Head
Head of the Department of Engineering Science and Mechanics
*Signatures are on file in the Graduate School
iii
Abstract
The focus on faster, smaller, and more diverse electronic devices has prompted
research into new materials. The search is on for semiconductor substrates with superior
electronic capabilities. Germanium has emerged as a semiconductor material capable of
producing desired performance. Current methods of thin-film deposition, once intended
for applications involving alternative substrates, must be altered to improve upon
germanium’s inherently unstable surface. This thesis investigates the adolescent
mechanism of self-assembly and chemical patterning on Ge(100). Our preliminary
research offers insight into applicable techniques for depositing stabilizing and
chemically functionalized, patterned adsorbate layers on germanium. It is our belief that
the examination of sample preparation will lead to reproducible and cost-efficient
techniques for use in industry. This dissertation will address possible candidates for
successful monolayer deposition on germanium, and the appearance of and possible
solutions to additional limitations.
Germanium suffers from an inherently unstable surface, a result of its native
oxide layer. Until this time, instability has limited germanium’s capacity for integration
into electronic devices. Self-assembled monolayers (SAMs) offer a method for
controlling interfacial properties, through formation of highly ordered, single-molecule
thick structures on surfaces, transferring their chemical attributes to the interface.
Successful deposition of SAMs on germanium can address its intrinsic limitations.
Through manipulation of SAM deposition techniques, used primarily for patterning
alkanethiols on gold, formation of high-quality monolayers becomes possible, improving
the surface stability of germanium.
iv
High-dielectric materials, due to their ability to resist leakage resulting from
electron tunneling, are necessary for producing functional components of nanoscale
dimensions. These materials may be limited in their ability to interact with relevant
substrates, making device fabrication difficult. Chemical patterning techniques exhibit
the ability to place an array of species selectively, otherwise limited by surface diffusion,
bonding capacity, etc., onto semiconductor surfaces. This dissertation describes the
utilization and manipulation of soft lithography methods for successful patterning on
germanium.
v
TABLE OF CONTENTS
List of Figures ............................................................................................................. vii
List of Abbreviations ................................................................................................. xii
Acknowledgments ...................................................................................................... xv
Chapter 1 Introduction.............................................................................................. ..1
1.1 Overview .................................................................................................................... ..1 1.2 Self-Assembled Monolayers ...................................................................................... ..3 1.3 Contact Angle ............................................................................................................ ..7
1.4 Chemical Patterning ................................................................................................... 13 1.5 X-Ray Photoelectron Spectroscopy ........................................................................... 18
Chapter 2 Self-Assembly on Germanium(100) ....................................................... 21
2.1 Overview .................................................................................................................... 21 2.2 Previous Work on Germanium ................................................................................... 23 2.3 Experimental Methods ............................................................................................... 24
2.3.1 Materials .......................................................................................................... 25 2.3.2 Fabrication of Monolayers .............................................................................. 25 2.3.3 Volcano, Kinetics, and Stability Measurements ............................................. 26 2.3.4 Contact Angle Measurements ......................................................................... 27 2.3.5 X-Ray Photoelectron Spectroscopy Analysis .................................................. 28
2.4 Results and Discussion ............................................................................................... 28
2.4.1 Mixed-Solvent System .................................................................................... 30
2.4.2 Deposition Kinetics ......................................................................................... 33
2.4.3 Long Term Stability ........................................................................................ 34
2.4.4 X-Ray Photoelectron Spectroscopy Analysis .................................................. 36 2.5 Conclusions ................................................................................................................ 40
Chapter 3 Chemical Patterning on Germanium(100) ............................................ 41
3.1 Overview .................................................................................................................... 41 3.2 Previous Work: Chemical Patterning and Thin Film Deposition .............................. 41
3.2.1 Vapor Deposition and Etching ........................................................................ 42 3.2.2 Additional Patterning Techniques for Semiconductors ................................... 43 3.2.3 Microcontact Printing on Germanium ............................................................. 45
3.3 Experimental Methods ............................................................................................... 46 3.3.1 Materials .......................................................................................................... 46 3.3.2 Stamp Preparation ........................................................................................... 47 3.3.3 Sample Preparation ......................................................................................... 48
3.3.4 Stamping Procedure ........................................................................................ 48 3.3.5 Analysis Techniques (XPS, SEM) .................................................................. 49
vi
3.4 Results and Discussion ............................................................................................... 50 3.4.1 Technique Progression .................................................................................... 50 3.4.2 X-Ray Photoelectron Spectroscopy Analysis .................................................. 52
3.4.3 Scanning Electron Microscopy Analysis ........................................................ 54 3.5 Conclusions ................................................................................................................ 58
Chapter 4 Conclusions and Future Prospects ......................................................... 59
4.1 Review ....................................................................................................................... 59 4.2 Future Work ............................................................................................................... 60
4.2.1 Chemicals, Solvents, and Crystalline Planes ................................................... 61
4.2.2 Extending Chemical Patterning ....................................................................... 62 4.2.3 Infrared Reflection Absorption Spectroscopy (IRRAS) .................................. 63
4.3 Conclusions and Final Thoughts ............................................................................... 67
Appendix 1 Non-Technical Abstract .............................................................................. 71
A.1 Non-Technical Abstract ............................................................................................ 71
References ................................................................................................................... 73
vii
List of Figures
Figure 1.1. Schematic of a n-alkanethiol (ALK) SAM on a Au{111} substrate. A. A
side view along the nearest neighbor direction showing the 30° tilt to
maximize the van der Waals interactions. B. Top down view showing the
unit cell of the Au substrate, as well as the (√3 × √3)R30° unit cell and the
c(4 × 2) superlattice. Shading highlights the c(4 × 2) structure. C. A single
surface-bound all-trans ALK molecule in which θ, ψ, and φ represent tilt,
twist, and azimuth angles of the chain, respectively.38
...…………………...6
Figure 1.2. Scanning tunneling microscopy images of a C12 SAM on a Au{111}
substrate acquired with a Vsample = -1.0 V and a Itunnel = 1.0 pA. The red,
yellow, and orange arrows highlight Au substrate step edges, Au substrate
vacancy islands, and molecular domain boundaries, respectively.
The circular protrusions correspond to single C12 molecules.38
…………...8
Figure 1.3. Cross-sectional schematic of a liquid droplet on a solid. The arrows
represent the location and direction of the surface tension forces associated
with the three interfaces. θC defines the contact angle…..............................11
Figure 1.4. A. Image representing contact angle measurements by the tilted wafer
method. The advancing angle is measured from the leading edge of the drop
(down hill side) and is denoted θA. The lagging edge produces the receding
angle, θR. The sliding angle, θS, is defined as the degree of tilt directly
before sliding motion occurs.47
B. Schematic depicting the addition and
reduction of drop size for an increase and decrease in liquid volume,
respectively………………………………………………………………...11
Figure 1.5. Image identifying the measuring parameters of a contact angle. The AB line
segment stretches adjacent and parallel to the surface from the point of
contact between the drop edge and the surface on the left and right side of
the drop. The BC segment begins at the right-most point of the liquid-solid
interface and extends tangentially to the drop’s surface at that point. The
angle of intersection between these two line segments is defined as the
contact angle.47
……………………………………………………………..14
viii
Figure1.6. Schematic showing basic stamping procedure for soft lithography
techniques. In all methods, an etched silicon wafer acts as a master key for
producing relief patterns in PDMS stamps. PDMS is placed in contact with
the master to assume its shape. The patterned PDMS stamp is peeled from
the wafer. It is then soaked in an “inking” solution. A. Microcontact
printing. The stamp is mechanically pressed upon the Au substrate. After
removal, the substrate will have a pattern of SAMs reproducing that of the
stamp. B. Microdisplacement printing. The inked stamp is mechanically
pressed on a Au substrate, pretreated with a labile SAM. The stamp
displaces the labile SAM in regions of contact, producing a fully covered
surface of labile SAMs, inlaid with inked monolayers. C. Microcontact
insertion printing. A non-labile SAM is deposited on a Au substrate.
Experimental variables control the physical characteristics of the preformed
monolayer. Upon contact, the inked stamp deposits inked monolayers into
defect sites.38
……………………………………………………………….16
Figure1.7. X-ray photoelectron survey spectrum of an AD SAM deposited from
solution. Note that every peak is directly associated with an atomic orbital.
Also, the Au 4d peak has a doublet resulting from spin-state splitting
induced by ionization.38
……………………………………………………20
Figure 2.1. The dynamic surface of Ge(100). Illustration demonstrates the degradation
of surface hydrophobicity as a result of oxide solubility. The first two drops
represent a non-wetted surface caused by the presence of a germanium oxide
layer. The third drop encountered the footprints of the previous two and
expresses the true wetting ability of the germanium surface. The first drops
actively removed the oxide layer and exposed the bare Ge(100) surface.
The last image was taken on a surface exposed to a 1:1 EtOH/H2O solution
overnight, further displaying the solubility of the oxide layer.64
…………..22
Figure 2.2. Step-by-step method of measuring contact angles by the dynamic sessile
drop method. A. Syringe is located over an un-tested region. B. A 5 μL drop
is deposited onto the substrate surface. C. The syringe is carefully centered
within the drop. D. The drop volume is increased to 10 μL, and the
advancing angle measurement is taken. E. The drop volume is increased
to15 μL (or until the wetted surface is increased). F. The drop volume is
decreased to 10 μL; the receding angle is measured.47
…………………….29
ix
Figure 2.3. The importance of solubility on direct SAM deposition on Ge(100).
Germanium oxide is soluble in water, while the alkanethiols are soluble in
alcohols. A water/alcohol mixture can solvate both the germanium oxide
and enough alkanethiol to form full-coverage SAMs. The unmixed solvents
(and exposure in a stepwise fashion) do not produce SAMs.25
……………31
Figure 2.4. Dependence of SAM quality on mixed solvent ratio. The appropriate
quantity of C12 is dissolved in the ethanol fraction, and then diluted with
water to the selected volume and ratio. Methanol yields a wider window of
solvent composition than ethanol. Films with an advancing contact angle
greater than 100° are considered to be of high quality. t-Butanol results are
not shown, as no ratio resulted in complete films.25
………………………32
Figure 2.5. Kinetics of SAM formation on Ge(100). Deposition is from 0.5 mM C12
solution in 1:1 ethanol/water. Advancing water contact angles are shown on
the logarithmic scale for clarity. The period of most dramatic change occurs
between 10 and 1000 min., and run-to-run variation occurs in this
region.25
…………………………………………………………………….35
Figure 2.6. Advancing contact angles collected after days of exposing C12 SAMs on
Ge(100) to the ambient environment, with samples either exposed to or
protected from light. There was no significant difference between the two
series.25
……………………………………………………………………..37
Figure 2.7. High-resolution XPS spectra collected for the regions specific to the C 1s,
S 2p, Ge 3d, O 1s, and Ge 2p. Each region is shown with its own arbitrary
intensity scale. A. Adventitious carbon (black trace) is effectively
eliminated by treatment with the water/ethanol (blue trace), enabling a
high-purity C12 SAM to form (red trace). B. The C12 monolayer gives the
expected response in the S 2p region, while the as-received and the cleaned
wafers do not show the presence of any sulfur. C. Oxidized species (red
trace) are removed readily by the water/ethanol treatment, leaving high
intensity peaks attributed to the bulk germanium crystal (blue trace). The
intensity is attenuated by the presence of the SAM (red trace). D. The Ge 3d
spectra provide information complimentary to that of the Ge 2p spectra.
Water treatment again eliminates the signal from the oxidized Ge species
(blue trace). The Ge 3d doublet appears shifted slightly to a lower binding
energy. The intensity is again attenuated by the presence of the monolayer
(red trace). E. The majority of the oxygen is attributed to the native
germanium oxide; after removal or SAM formation, remaining oxygen is
attributed to residual alcohols or oxide reformation at defect sites before the
samples were loaded into the XPS vacuum chamber.25
……………………38
x
Figure 3.1. Schematic comparing standard microcontact printing (a) and submerged
microcontact printing (b). A. PDMS stamp is produced through introduction
to an etched silicon wafer. B. The stamp is saturated by an ethanolic C12
solution. C. In standard microcontact printing the stamp is mechanically
pressed against the substrate under ambient conditions. Our method
involved contact between stamp and germanium in the presence of a water
medium, so as to eliminate the native oxide. D. Adsorbate pattern is
deposited onto substrate in atmospheric air (a), and while submerged in
water (b).106
………………………………………………………..……….51
Figure 3.2. High-resolution XPS spectra collected for the region specific to the C 1s
orbital. The adventitious carbon of the as-received sample (red trace), is
effectively removed by rinsing with water/ethanol (green trace), allowing a
high quality C12 monolayer (purple trace) to form. Our method of
submerged microcontact printing on germanium produces a monolayer of
50% coverage (black trace).64
…………...………………………………...53
Figure 3.3. Scanning electron microscopy (SEM) images of C12 monolayers, patterned
on germanium (100). Images represent a single patterned surface at varying
resolutions.64
…………………………………………...…………………..55
Figure 3.4. Scanning electron microscopy (SEM) image of C12 monolayer, patterned on
germanium (100), backfilled with MUDA. The inversion of intensity, in
comparison to the untreated C12 monolayer, is indicative of MUDA SAM
formation. Successful backfilling represents our ability to control the
chemistry of both the patterned and non-patterned regions. Retained edge
fidelity of the patterned squares shows the non-damaging nature of the
backfilling procedure and the stability of the monolayers. A lack of C12
monolayer coverage results in limited contrast between regions.64
……….56
Figure 3.5. Scanning electron microscopy (SEM) image of C12 monolayer, patterned on
germanium (100), backfilled with MUDA. This image is a contrast
enhanced version of Figure 3.4, so as to simplify feature decipherability.
The inversion of intensity, in comparison to the untreated C12 monolayer, is
indicative of MUDA SAM formation. Successful backfilling represents our
ability to control the chemistry of both the patterned and non-patterned
regions. Retained edge fidelity of the patterned squares shows the
non-damaging nature of the backfilling procedure and the stability of the
monolayers. A lack of C12 monolayer coverage results in limited contrast
between regions.64
……………………………………………………...….57
xi
Figure 4.1. Stretching and bending vibrational modes of a methylene (–CH2–) group.
38 …………………………………………………………………….64
Figure 4.2. Infrared reflection absorption spectroscopy (IRRAS) spectra of a C12 SAM
on a Au{111} substrate, showing higher energy stretching modes (A), and
lower energy bending and wagging modes (B).38
…………………………65
Figure 4.3. Schematic of the attenuated total reflectance assembly. An absorbing
medium is placed in intimate contact with an infrared transparent crystal, the
internal reflection element. The angle of incidence is set such that it
exceeds the critical angle, ensuring total internal reflection. The crystal
edges are cut as to allow entrance and emission of incident
beam.116
……………………………………………………………………68
Figure 4.4. Schematic depicting the formation of an evanescent wave as a result of total
internal reflection. Upon contact with the IRE boundary, an electric field, in
the form of an evanescent wave will develop, as described by Schrödinger’s
wave function. In the presence of an absorbing medium, the incident beam
will experience a reduction in intensity, corresponding to the medium’s
absorption of the evanescent wave, making absorption spectra collection
possible.117
………………………………………………………………...69
xii
List of Abbreviations
AD
ALK
C2
C12
C18
MUDA
OTS
PDMS
1-Adamantanethiol
Alkanethiol
n-Dithiol
n-Dodecanethiol
n-Octadecanethiol
11-Mercaptoundecanoic Acid
Octadecyltrichlorosilane
Polydimethylsiloxane
CVD
CVM
DPN
LPCVD
MOCVD
PLD
PVD
RIE
UHV-CVD
Chemical Vapor Deposition
Chemical Vapor Machining
Dip Pen Nanolithography
Low-Pressure Chemical Vapor Deposition
Metal-Organic Chemical Vapor Deposition
Pulsed-Laser Deposition
Physical Vapor Deposition
Reactive Ion Etching
Ultrahigh Vacuum Chemical Vapor Deposition
Molecules
Patterning Techniques
xiii
AFM
ATR
FESEM
IRRAS
XPS
Atomic Force Microscopy
Attenuated Total Reflectance
Field Emission Scanning Electron Microscopy
Infrared Reflection Absorption Spectroscopy
X-Ray Photoelectron Spectroscopy
BE
GL
IRE
Itunnel
KE
SAM
UHV
Vsample
Binding Energy
Gaussian-Lorentzian
Internal Reflection Element
Tunneling Current Set Point
Kinetic Energy
Self-Assembled Monolayer
Ultrahigh Vacuum
Applied Sample Bias
Other Acronyms
Characterization Tools
SμCP
μCP
μCIP
μDP
Submerged Microcontact Printing
Microcontact Printing
Microcontact Insertion Printing
Microdisplacement Printing
xiv
γLV
γSL
γSV
θA
θC
θR
θS
θγ
Surface Tension at Liquid-Vapor Interface
Surface Tension at Solid-Liquid Interface
Surface Tension at Solid-Vapor Interface
Advancing Contact Angle
Equilibrium Contact Angle
Receding Contact Angle
Sliding Angle
Young’s Contact Angle
xv
Acknowledgments
Writing this dissertation has been an interesting experience, to say the least. I
would not recommend condensing the thesis writing adventure into a few weeks, but I
have discovered it can be done. First, I would like to thank my thesis advisor, Professor
Paul S. Weiss. Not only did you afford me the opportunity to explore graduate studies,
you continue to offer me every chance at success in the field. I appreciate your
understanding demeanor concerning my future plans and your willingness to assist in my
endeavors. I would also like to thank you for the indirect supply of motivation necessary
for me to start (and quickly finish) this thesis. Professor Mark W. Horn, thank you for
taking me on as your advisee. Without your help, I would not have been able to pursue a
degree in the engineering field. Additionally, I would like to thank Professor Tony J.
Huang for agreeing to serve on my thesis committee. Working predominately within the
chemistry department, I knew few engineering science professors, and am glad you
agreed to finalize my committee.
I would like to thank the National Science Foundation, the National Science
Foundation funded Center for Nanoscale Science, the Army Research Office, the Kavli
Foundation, and Penn State’s National Nanotechnology Infrastructure Network for their
financial support of this work. Additionally, the Penn State University Materials
Characterization Laboratory, and especially Dr. Thomas A. Daniel are gratefully
acknowledged for data collection and XPS analysis.
xvi
The completion of this thesis would not have been possible without the assistance
of several other individuals. First, I would like to thank Dr. Steve Stranick. Unsure of
my future plans, you encouraged and enabled me to pursue the field of research. My time
at N.I.S.T. was informative and enjoyable. Furthermore, without your guidance and
recommendations I would have never considered or been able to achieve a graduate
degree. Your ability and patience concerning the translation of research concepts into
layman’s terms has proved incalculable. Additionally, I think I learned just as much on
the rides to and from work as I did in the lab. The sports conversations were entertaining,
but the amount of information I gained concerning random issues was phenomenal. I
think of you not only as a mentor, but as a friend, thank you.
I would also like to extend a big thank you to all of the Weiss group members,
past and present. Without your assistance, concerning a number of issues, this would not
have been possible. I would especially like to thank Nate Hohman. He guided my
research, encouraged me to learn as much as I could, and without his clarification and
editing, I doubt this dissertation would have been legible.
Condensing the next few people into one paragraph is almost an injustice. My
family. Mom, Dad, Laura, Kevin, and Heather, although I was reluctant to explain, and
you would have failed to understand anyway, you always inquired about my progress and
offered encouragement. To my extended family (Uncle Bill, Aunt Johnna, Joe, and Nick,
consider yourselves included, I appreciate all of your support and maybe someday I will
get that PhD), thanks for the prayers and the push to continue through indecision. Rick
and Jen, if you lived more than 10 minutes away, there is a good chance that I would be
writing this paper in smelly clothing. Hell, I might not be writing it at all as a result of
xvii
medical complications related to malnutrition. Furthermore, without the interaction of
you and your slightly crazy neighbors (I plan to burn that wood as quickly as possible, so
we can look for more, if they even let us see each other), I would have been
institutionalized, as a result of my inability to excel socially. To Lilly, Olivia, and
Connor; thanks for the comic relief. Without you little ones, boredom would have
ensued. I am glad that my graduate work enabled me to watch you guys grow up.
Maybe, before you guys get too big, I will use you to find myself a girlfriend, seeing as
though your mother has been incapable of doing so. I love you. To the Zeigler crew,
Trigger says thanks for the good times around the fire, and the ridiculous hours spent
sprinting around the mountain in an attempt to keep up with the “Master.” Make sure to
pick up any “road closed” signs that you see and we will drop them off on our way to
visit Hook and the gang. It is alright to eat the chitlins but stay away from the pepper jack
cheese, and if you do either, remember to pack the paper towels. Never forget that every
dog has its day, and of course, to all my blood brothers, Voodoo! To all you other fools
out there, I call my friends, thanks for using my apartment as a gateway to explore the
nightlife of state college. Drinking beers with you guys is just good old fashion fun.
Finally, to the big guy upstairs, I am fairly confident that it was you who kept me sane
throughout all this. I appreciate it.
Chapter 1
Introduction
1.1 Overview
The self- and directed assembly of molecularly-thick films constitute a powerful
tool for use in the areas of nanotechnology, molecular electronics, biochemistry, and
bioengineering.1-8
Extensive research has been conducted to identify quality candidates
for controlling interface properties and for the patterning of technologically important
substrates. Gold and silicon have emerged as important materials for their stability,
scientifically functional characteristics, and molecular compatibility. For this reason,
they have been extensively studied and are relatively well understood. The ability to
manipulate such substrates and their interfacial properties is key for the advancement of
technology as is the incorporation of new materials and novel methods.
Presently, Si remains the most important technological semiconductor for its
ability to produce a quality dielectric oxide layer.6, 7
The SiO2 oxide layer passivates the
surface of silicon and stabilizes the interface between Si and SiO2.7 The density of
devices per unit area has increased steadily since the development of the integrated
circuit, a well-publicized trend known as Moore’s law.9 Recently, the demand for a shift
from macroscale to nanoscale electronics has driven a search for new strategies for
electronic devices. On such small scales, leakage currents through dielectrics (a result of
electron tunneling between features at small length scales) wreak havoc on performance.7
For this reason, materials with dielectric constants superior to that of SiO2 are being
sought after. With the advent of dielectric films other than SiO2, semiconductors with
better electrical characteristics than those of silicon become preferable.7 Group IV
2
semiconductors are the most obvious choice because of their elemental simplicity and
their potential for applications in integrated circuits, solar cells, infrared (IR) detectors,
and chemical sensors, amongst others.7
The intrinsic properties of germanium have made it a potential material for a new
generation of fast electronics. Germanium offers excellent intrinsic electrical properties,
specifically higher electron and hole mobility than silicon, and a smaller bandgap.7, 8, 10
Unfortunately, the native germanium oxide is a poor dielectric making it difficult to apply
germanium in applications where silicon is poorly suited. Passivation strategies to
control the reactivity of silicon are already well-understood, and similar strategies must
be developed for germanium if we are to take advantage of its electronic characteristics.7
The electrical properties of high-dielectric film/substrate interfaces are dependent upon
deposition process, deposition parameters, and pre-deposition surface treatments.8
Engineering and characterization of interfacial properties are critical for assuring optimal
electronic performance.8 With proper experimentation and manipulation, germanium can
outperform its predecessors and be used to push the limits of technology.
This dissertation explores the use of self-assembly and chemistry to enable
established patterning techniques to extend beyond standard materials and onto the
relatively less explored semiconductor substrate Ge(100). I will discuss the surface
chemistry of the germanium surface and oxide, and the development of new methods for
both controlling that chemistry and extending chemical patterning to germanium
surfaces. A number of techniques were used both to develop and to characterize
patterned surfaces. For this reason, the remainder of this chapter will provide an
introduction to the most important concepts and analysis methods applied in this paper.
3
A description of self-assembled monolayers, contact angle procedure, chemical
patterning and x-ray photoelectron spectroscopy follow.
1.2 Self-Assembled Monolayers
When an organic molecule adsorbs on a metal or metal oxide surface, the free
energy of the interface is lowered.11
Due to the energetically favorable nature of the
interaction, metallic surfaces tend to accommodate nearly any organic adsorbate. By
carefully controlling the surface and the organic adsorbate, highly ordered structures may
be spontaneously generated from a disordered solution, in a process known as
self-assembly. These molecules arrange on the surface in accordance to their physical
and chemical properties, which can be tailored for this purpose.12
Self-assembled
monolayers (SAMs) offer a method for controlling interfacial properties by forming
highly ordered, single-molecule-thick structures on surfaces, transferring its chemical
properties to the interface.12-24
These SAMs are usually organic self-assemblies formed
through adsorption of molecules from the gas or liquid phases.12
The molecules that
form SAMs tend to have a strong affinity for the surface, enabling them to displace
pre-adsorbed species and to form well-ordered, chemically functional monolayers.
Typically, all reactive substrate sites will be occupied within seconds of exposure to the
molecular solution. This rapid assembly accounts for the majority of adsorption.12
After
the initial assembly, a re-ordering process occurs on a timescale of hours. During this
time, the monolayer tends to maximize adsorbate density, while improving nanoscale
order so as to eliminate defects.12
For a millimolar solution under ambient conditions the
initial assembly takes no more than a few minutes and the re-ordering, no more than a
4
few hours. However, variables in preparation, such as molecular concentration and
temperature, can affect the assembly process.12
A wide range of substrates, including noble metals, base metals, insulators, alloys,
and semiconductors have been used for SAM preparation.12, 25-32
Substrate materials are
found in many forms, including planar surfaces (glass, metal films) and various
three-dimensional nanostructures (colloids, nanocrystals, nanorods). The most commonly
studied class of SAMs are those formed from the deposition of alkanethiols on
Au{111}.15, 16, 33, 34
The strong affinity of thiols for Au allow it to form exceptionally
ordered monolayers with diverse exposed functionality. Gold is a particularly attractive
material for a number of spectroscopic and lithographic techniques, as it does not oxidize
in air and shows limited catalytic activity.15
The stable and well-ordered SAMs of alkanethiols on gold constitute a model
system for self-assembly. The interaction between Au{111} and thiol produce strong
stabilizing bonds with an enthalpy of ~44 kcal/(per unit) mol.19
The thiol headgroup
interacts with the gold substrate to produce a Au-S bond, and results in loss of the
hydrogen atom, presumably in the form of dihydrogen.12
The exothermic bond formation
drives SAM assembly, and enables secondary interactions between adjacent alkyl chains.
The methylene groups associated with each alkyl chain contribute an additional ~1
kcal/(per unit) mol of stabilizing van der Waals interactions.16
An alkanethiolate SAM is well known to form a hexagonal (√3 × √3)R30° unit
cell at the Au{111} interface, with a nearest neighbor spacing of ~4.97Å.18, 19
The
interactions between alkyl chains tend to drive a c(4×2) superlattice structure
representative of the methyl groups in relation to the Au surface.12, 18, 19, 33
5
The lateral interactions between these chains drive the ordering of the SAM, and
for that reason longer carbon chains show more order as a result of increased van der
Waals interactions.35
In the case of alkanethiols on Au, the molecules exhibit a molecular
tilt of ~30° with respect to the surface normal.34, 36
Figure 1.1 shows the formation of the
alkanethiol unit cell and the c(4×2) superlattice; as well as the orientation of the alkyl
chains in relation to the Au substrate.
The stability and order of SAMs is well documented, but defects do exist.
Disruptions in the monolayer can be caused by defects within the substrate and/or
inconsistencies in the ordering process of assembly. Substrate defects are natural features
of even the most carefully prepared metallic surface, with common features of atomic
steps, vacancy islands, and intergrain boundaries.17
Atomic steps and grain boundaries
are intrinsic to crystalline substrates, whereas vacancy islands arise from the chemistry of
self-assembly. A typical Au{111} surface exhibits a (23 × √3) reconstruction, and
contains a greater density of atoms then the ideal bulk-constructed {111} crystal face.37
When thiols bond to the surface during the self-assembly process, they release the
reconstruction and reduce atomic density of the substrate surface. The healing
mechanism of the surface comes in the form of a single atom vacancy, which will
typically diffuse until it collides with another vacancy, resulting in the nucleation and
growth of a “vacancy island.”37
All three substrate defects result in uneven topography,
which is reproduced by the monolayer. Self-assembly defects are present in the form of
compositional heterogeneity and domain boundaries. Self-assembly is a thermodynamic
process governed by competitive binding.12
If adsorbates exhibiting strong interactions
contact the substrate, they can infuse themselves into an otherwise homogenous
6
Figure 1.1: Schematic of a n-alkanethiol (ALK) SAM on a Au{111} substrate. A. A side
view along the nearest neighbor direction showing the 30° tilt to maximize the van der
Waals interactions. B. Top down view showing the unit cell of the Au substrate, as well
as the (√3 × √3)R30° unit cell and the c(4 × 2) superlattice. Shading highlights the
c(4 × 2) structure. C. A single surface-bound all-trans ALK molecule in which θ, ψ, and
φ represent tilt, twist, and azimuth angles of the chain, respectively.38
7
monolayer. A domain boundary is characterized by the intersection of alkyl chains with
differing azimuthal tilt angles or offset lattices of attachment. As a result of the chain’s
dependence on the Au-S packing structure, they may assemble differently, causing
multiple orientations on the surface. The red, yellow, and orange arrows in Figure 1.2
depict step edges, vacancy islands, and domain boundaries within a SAM, respectively.
Manipulations of SAMs are far from limited. For example, molecules may be
inserted into defects, predominately at step edges and domain boundaries, a technique
called insertion.12
Also, monolayers can be produced such that varying lengths of alkyl
chains are codeposited simultaneously on the surface. These mixed monolayers allow the
engineering of additional defect sites, many in the form of domain boundaries, where
useful chemistry techniques can be applied for the purpose of studying alternative
molecules. Self-assembled monolayers have proven to be a useful tool in the
development of molecular assemblies and offer a means for controlling the interfacial
properties of a metallic surface.
1.3 Contact Angle
The physical characteristics of a liquid droplet on the surface of a solid substrate
are dependent upon the chemical and physical interactions at the contact interface.
Through careful physical analysis of the droplet, some idea of the interfacial properties
can be gained. Contact angle measurement is a straightforward and accurate way of
measuring wettability, roughness, interfacial energy, and chemical composition of a
surface.39-41
For this reason, we use the measurement of contact angles as a method for
gaining insight into and analysis of surface behavior.
8
Figure 1.2: Scanning tunneling microscopy images of a C12 SAM on a Au{111}
substrate acquired with a Vsample = -1.0 V and a Itunnel = 1.0 pA. The red, yellow, and
orange arrows highlight Au substrate step edges, Au substrate vacancy islands, and
molecular domain boundaries, respectively. The circular protrusions correspond to single
C12 molecules.38
9
Three interfaces exist at the point of contact between liquid and solid. The
solid-liquid interface, solid-vapor interface, and liquid-vapor interface all express unique
interfacial properties. The intersection of the three interfaces will extend along the
perimeter of the droplet, and is named the triple or pinning line.42
The contact angle is
measured at a point where the pinning line is reduced to a single point, as is shown in the
cross-sectional graph of Figure 1.3, and is defined as “the angle between the solid
surface and a tangent drawn to the liquid surface at the point of contact with the solid.” 43
The relationship between the three interfaces is expressed by Young’s equation:
,cos SLSVLV
where γ is the surface tension at the interface and θγ is Young’s angle.41
The subscripts
L,V, and S stand for liquid, vapor, and surface, respectively. Surface tension is a physical
value representing the amount of work per area necessary to expand a surface.44
The
surface tension for a given material will actively resist expansion. Because every
compound exhibits a distinct value for surface tension, the contact angle is a variable
dependent upon the interaction between the involved species.
Young’s contact angle, also referred to as the equilibrium contact angle, is a
product of the interfacial tensions. The equilibrium contact angle occurs when the
chemical potential of the three phases is equal and should be singularly constant for a
given system. The angle measured by contact angle techniques is rarely consistent with
this equilibrium value, as a result of surface impurities.40
In fact, stable drops may be
observed over a range of contact angles. For this reason, contact angles are often defined
in terms of two quantities, the advancing angle and receding angle. The advancing angle
is defined as the angle associated with an increase in wetted area.40
Depending on the
measurement technique, this could come by way of droplet perimeter expansion, or the
(1.1)
10
leading edge of a mobile droplet. Likewise, the receding angle is measured for recession
in wetted area and can be observed for the lagging edge of moving liquid, or as the drop
is shrinking in size. The numerical difference between these two values is known as
contact angle hysteresis.40
Figure 1.4 shows advancing and receding angles as defined
for both the tilted wafer method and dynamic sessile drop method.
Hysteresis results from chemical heterogeneity and surface roughness.45
Surface
roughness was first addressed by Robert Wenzel when he suggested that the presence of
surface heterogeneity caused the discrepancy between actual surface area and geometrical
surface area. Using his modified Young’s equation to account for this difference, he
proposed that the ratio of actual surface area to geometric surface area be classified as the
roughness factor, r.46
Wenzel’s Equation is expressed:
where γ and θ are defined in accordance with Young’s equation and r is the roughness
factor. This roughness factor accounts for real-world applications where the substrate
surface invariably fails to be completely flat. Experimental data have shown that
roughening a surface will cause the advancing angle to increase and the receding angle to
decrease. In some cases, these results disagree with Wenzel’s equation.40
In either case,
a change in angle due to surface roughness is largely a function of surface geometry and
is not affected by the interfacial tensions.
The existence of hysteresis suggests the presence of surface roughness and
chemical heterogeneity. It is impossible to make conclusions concerning the
contributions of either source as they are indistinguishable from the data analysis alone.
Only when one source can be completely eliminated can assumptions be made.
(1.2) γLV cosθγ =( γSV - γSL) r,
11
Figure 1.3: Cross-sectional schematic of a liquid droplet on a solid. The arrows
represent the location and direction of the surface tension forces associated with the three
interfaces. θC defines the contact angle.
Figure 1.4: A. Image representing contact angle measurements by the tilted wafer
method. The advancing angle is measured from the leading edge of the drop (down hill
side) and is denoted θA. The lagging edge produces the receding angle, θR. The sliding
angle, θS, is defined as the degree of tilt directly before sliding motion occurs.47
B. Schematic depicting the addition and reduction of drop size for an increase and
decrease in liquid volume, respectively.
Tilted wafer
θA
θR
θS
A B
12
For example, if a surface is known to be consistently flat, then the effects of surface
roughness are negligible.45
In this case, the presence of hysteresis is the result of
chemical heterogeneity. Furthermore, if it can be determined that hysteresis is caused
predominately by chemical impurities, then Young’s equation can still be applied to
make some statement concerning surface tension.
There are several methods to measure contact angles. Among the most frequently
used are the Wilhelmy method, tilted wafer method, and sessile drop method. The
Wilhelmy method is a descendent of the Wilhelmy plate method, used to measure a
liquid’s surface tension. In the Wilhelmy plate method, a plate of known dimensions and
buoyancy is exposed to a liquid bath. The plate is suspended from a microbalance and the
change in mass after exposure is recorded.48
Through modification of Young’s equation,
the advancing and receding angles can be calculated. The tilted wafer method is a
dynamic drop method that involves placing a droplet on an angled substrate.49
The
substrate is tilted such that the droplet is on the verge of sliding motion. The leading (or
down hill side) of the droplet is measured and recorded as the advancing angle. The
trailing edge is measured to gain the receding angle data.49
For the measurements
discussed in this paper, the dynamic sessile drop method was applied. The dynamic
sessile drop method utilizes a microsyringe to deposit liquid droplets onto a solid surface.
Through addition and subtraction of liquid, the droplet will expand and contract,
respectively. The advancing angle is measured while the droplet is in the “filling” mode
and the receding angle is recorded during the “emptying” mode.49
It is generally
accepted that an approximate value of the equilibrium angle can be achieved by
13
averaging the advancing and receding angles. Figure 1.5 defines the contact angle,
represented by B.
The evolution of surface tension, and ergo contact angles, associated with a
change in chemical composition, is particularly relevant for the work discussed in this
dissertation. Here, contact angle measurements were used to measure changes in
wettability to track a chemical reaction. We compared the contact angles from a bare,
polished Ge(100) wafer to those exposed to various molecular solutions, in hopes of
making some statement about the deposition of a monolayer onto the germanium surface.
After exposure to dodecanethiol solutions, the contact angle of the surface increased
dramatically, consistent with compositional changes in surface chemistry.
1.4 Chemical Patterning
The relevance of chemical patterning is ever increasing. The films produced by
patterning techniques are applicable to numerous scientific fields including biological
surface science, bio-medicine, and molecular electronics.1, 3-5, 13, 50-53
There are numerous
methods for chemical patterning, many of which fall under the category of soft
lithography techniques. Soft lithography is a set of techniques where patterns are
transferred to substrates in the absence of energetic beams, by means of soft and flexible
materials.3, 54
Unlike hard lithography and other patterning techniques, soft lithography
does not compromise the integrity of the substrate.3 Soft lithography has exhibited the
ability to comply with a large number of adsorbates and substrates. It is utilized in
situations where other lithography techniques are incapable of performing, such as
patterning applications on curved surfaces and biologically compatible surfaces with
14
Figure 1.5: Image identifying the measuring parameters of a contact angle. The AB line
segment stretches adjacent and parallel to the surface from the point of contact between
the drop edge and the surface on the left and right side of the drop. The BC segment
begins at the right-most point of the liquid-solid interface and extends tangentially to the
drop’s surface at that point. The angle of intersection between these two line segments is
defined as the contact angle.47
A B
C
15
feature characteristics on the scale of hundreds of nanometers.54
Of the many soft
lithography techniques, microcontact printing (μCP) emerged as one of the most
successful. The typical method for transferring adsorbates to the surface includes a
polydimethylsiloxane (PDMS) stamp, which has been “inked” with a molecular
adsorbate.3, 50
The elastomeric stamp is designed with surface reliefs that are a mirror
image of the desired deposition pattern making selective placement possible.55
The
PDMS stamp is soaked in an inking solution, typically a millimolar solution of
alkanethiol dissolved in ethanol. After being blow dried, the stamp is pressed
mechanically onto the substrate surface. A commonly used substrate of choice is
Au{111}, with alkanethiol inks for preparing patterned self-assembled monolayers in
contact zones. However, the stamp is capable of patterning onto any thiol-compatible
surface. Later in this thesis, a novel method for microcontact printing on Ge(100) will be
discussed.
During contact, alkanethiols transfer from the stamp to the substrate and
self-assemble according to the relief pattern of the stamp. Figure 1.6 shows a schematic
of the stamping process. The resulting SAM is sensitive to the mole fraction of the
inking solution, affording some control over the patterning process by varying the thiol
concentration.55-57
In fact, it has been reported in particular cases that chemically
patterned SAMs are functionally indistinguishable from SAMs produced by means of
solution deposition.57
The capabilities of a microcontact-printed surface are numerous. The covalent
assembly of the Au/thiol reaction allows for considerable control of the samples
16
Figure 1.6: Schematic showing basic stamping procedure for soft lithography
techniques. In all methods, an etched silicon wafer acts as a master key for producing
relief patterns in PDMS stamps. PDMS is placed in contact with the master to assume
its shape. The patterned PDMS stamp is peeled from the wafer. It is then soaked in an
“inking” solution. A. Microcontact printing. The stamp is mechanically pressed upon
the Au substrate. After removal, the substrate will have a pattern of SAMs reproducing
that of the stamp. B. Microdisplacement printing. The inked stamp is mechanically
pressed on a Au substrate, pretreated with a labile SAM. The stamp displaces the
labile SAM in regions of contact, producing a fully covered surface of labile SAMs,
inlaid with inked monolayers. C. Microcontact insertion printing. A non-labile SAM is
deposited on a Au substrate. Experimental variables control the physical
characteristics of the preformed monolayer. Upon contact, the inked stamp deposits
inked monolayers into defect sites.38
17
properties. The SAM will actively resist the advancement of etchants and shield its
supporting metal substrate from corrosion.3, 55, 58
The monolayer offers control over
surface characteristics including wettability and electrical conductivity.55, 58
Additional
highlights of microcontact printing include cost efficiency, capacity for large printing
areas, and the disproportionate relationship between the ease of preparation and the
quality of results. Soft lithography does have some negative aspects. Issues with surface
diffusion, resolution, and reproducibility have required further development of printing
mechanisms.59
Microdisplacement printing (μDP) utilizes a substrate with a pre-formed
monolayer of 1-adamantanethiolate (AD).59
The use of a pre-formed monolayer is
two-fold. First, its presence prevents the lateral diffusion of patterning solutions allowing
for a greater variety of applicable molecules. Secondly, AD forms a weakly bound SAM
on Au and is therefore labile. The labile nature of the molecule allows it to be displaced
through competitive adsorption. Therefore, microcontant printing techniques can be
applied to replace the labile SAM with a more desirable one.59
The resulting film is a
compilation of AD and patterning or “displacing” SAMs.
Another novel printing technique is microcontact insertion printing (μCIP). Like
microdisplacement printing, this method utilizes both microcontact techniques and a
pre-assembled monolayer.2 The composition of the patterned SAM can be controlled
through manipulation of stamping duration, concentration of molecules, and the quality
of the pre-assembled monolayer.2 Unlike microdisplacement printing, the pre-existing
SAM is not labile relative to exchange. Therefore, when the stamp contacts the surface,
18
the insertion sites are limited by defects. By manipulating the pre-existing film, we gain
control over the chemical environment and molecules can be isolated from one another in
separated defect sites.2
Section 1.5 X-Ray Photoelectron Spectroscopy
Although contact angle measurements are a quality diagnostic for identifying
some surface behavior, they do not provide detailed information about the chemistry of
the surface. In order to decipher what was occurring on the pre- and post-treated
germanium surface, we require a surface-selective and chemically specific analysis
technique. X-ray photoelectron spectroscopy (XPS) offers definitive evidence
concerning the chemical composition of the surface in question. To aid in the
understanding and interpretation of the XPS data present in this dissertation, we will
briefly address the principles of this spectroscopic technique.
X-ray photoelectron spectroscopy utilizes incident photons for the excitation and
emission of core electrons from surface atoms. Mono-energetic x-rays irradiate a solid
surface, and by means of the photoelectric effect, enable surface electrons to escape the
bulk in the form of photoelectrons.60
X-ray photoelectron spectroscopy is surface
selective, because photoelectrons in solids are characterized by a small mean free path,
and cannot reach the detector if they are generated in the bulk of the material. To improve
surface selectivity further, Mg Kα and Al Kα x-rays are typically used for their low
penetrating depths in metal (1-10 μm).60
Although electrons produced deeper then a few
tens of Ångstroms below the surface can be emitted, they will incur significant inelastic
energy loss. Those electrons produced within tens of Ångstroms of the surface can
19
escape without energy loss and provide the most useful data.60
The kinetic energy (KE)
of the photoelectron is given by:
,BEhKE
where hν is the energy of the incident photon, φ is the work function of the spectrometer,
and BE is the orbital’s binding energy. Binding energy is classified as the difference
between the initial and final energy states of the atom after emission. It is also defined as
a value proportional to the difference in energy between an electron’s orbital and the
Fermi level of the solid. Higher orbital electrons dwell closer to the Fermi level and thus
require less energy to break existing bonds.
The XPS instrument analyzes the emitted photoelectrons. The raw data translates
to a plot of the number or intensity of electrons per energy interval versus kinetic or
binding energy. Because every element has characteristic electronic orbitals and a related
variety of ionic states, it will thus display a distinctive spectrum of photoelectron
emission.60
Through analysis of a sample’s energy signature, the occupying elements can
be identified. Furthermore, for a sample of mixed composition, the spectrum will reveal
an approximate sum of peaks belonging to the individual elements.60
Their identities and
concentrations can be measured. Conveniently, spectra have already been measured for
many of the elements and some basic elemental combinations, simplifying the
identification of chemical unknowns. Figure 1.7 depicts an XPS survey spectrum. Note
that every peak correlates to a specific binding energy and atomic orbital. Also, note the
Au 4d peak(s). As the surface is excited by incident photons the atoms become ionized,
and illuminate the spin states of slightly different kinetic energy of the surface atoms.60
(1.3)
20
Figure 1.7: X-ray photoelectron survey spectrum of an AD SAM deposited from
solution. Note that every peak is directly associated with an atomic orbital. Also, the Au
4d peak has a doublet resulting from spin-state splitting induced by ionization.38
21
Chapter 2
Self-Assembly on Germanium(100)
2.1 Overview
The reign of silicon as the leading semiconductor material in electronic
applications may be nearing its end - a result of advances in dielectric film production.
Other group IV semiconductors, because of their ability to function in a wide range of
modern devices, provide the most capable replacements. Germanium has emerged as a
candidate for future generations of fast, electronic devices, due to its high mobility and
narrow bandgap.7, 8, 10
Germanium is presently found in few devices, a result of its poor native oxide.
This oxide layer is poorly passivating, thermally unstable, and water soluble.6, 25, 61, 62
When germanium is exposed to the atmosphere, a mixture of oxides, including GeO and
GeO2, form on its surface, in a process akin to oxidation of aluminum.6, 62, 63
GeO2 is an
inferior dielectric and is water soluble.6, 62
Both oxides suffer from instability. Further,
the interface between germanium and its oxide is amorphous, and exhibits a significant
amount of defects.62
These features have limited germanium’s utility and
interchangeability with silicon. Figure 2.1 details the presence and solubility of the
germanium oxide.
Techniques for the removal of germanium’s oxide layer have been developed.
Atomically clean surfaces have been achieved in ultrahigh vacuum (UHV) cleaning
processes, but wet chemical techniques in ambient conditions are required for industrial
purposes.6, 62, 63
Although dielectric films have been successfully deposited onto clean
22
Figure 2.1: The dynamic surface of Ge(100). Illustration demonstrates the degradation
of surface hydrophobicity as a result of oxide solubility. The first two drops represent a
non-wetted surface caused by the presence of a germanium oxide layer. The third drop
encountered the footprints of the previous two and expresses the true wetting ability of
the germanium surface. The first drops actively removed the oxide layer and exposed the
bare Ge(100) surface. The last image was taken on a surface exposed to a 1:1 EtOH/H2O
solution overnight, further displaying the solubility of the oxide layer.64
23
germanium surfaces, the interface characteristics of the interaction produce leaky and
undesirable devices.6 Until this point, it has been shown that removal of the oxide layer
alone is insufficient for quality electronic performance, and for this reason, the cultivation
of passivation techniques of germanium must be developed for production of functional
components.6
2.2 Previous Work on Germanium
Germanium passivation is a topic of growing interest. Strategies involving
hydrogen, halogen, sulfide, and alkyl passivation have been examined.25
As described in
chapter 1, alkanethiols have the ability to form well-ordered SAMs on a variety of
surfaces. Self-assembly on germanium will provide control of its interfacial properties
and improve surface passivation.6, 65-67
Cullen and co-workers demonstrated germanium alkylation by Grignard reaction
in the 1960s.68
Forming a Ge-C bond protects the surface from ambient conditions. The
surface was first chlorinated by exposure to hydrogen chloride and chlorine gases,
followed by ethyl magnesium bromide introduction.68
Later work focused on
halogenation and sulphidization. Surfaces of semi-conductors are highly reactive, and
oxidation occurs as the result of unsaturated dangling bonds. To stabilize the surface, Lu
introduced the substrate to HCl, which satisfied the reactive dangling bonds with
monochloride.69
Germanium(111) surfaces were shown to exhibit stable chlorinated
surfaces in ambient conditions by means of wet chemical exposure to HCl.69
Other
studies have investigated the use of sulfur as a passivating agent.70-72
The interaction of
24
sulfur and germanium breaks down surface dimers on the Ge substrate, allowing sulfur to
adsorb at the interface and form a (1 ×1) reconstruction.70-72
Germanium in various
oxidation states were shown to remain on the reconstructed surface, however, leading to
non-ideal surface properties. Recently, the deposition of alkyl chains on germanium has
re-emerged as a useful technique for producing ideal surface conditions. The direct Ge-S
bond leads to well-ordered monolayers giving good coverage and thermodynamic
stability up to 450K.6, 65, 73
The monolayer provides a means for surface passivation and
offers the ability to modify surface properties and to integrate biological functionalities
into solid-state devices. Current methods for SAM deposition on germanium involve
multi-step processes, where oxide removal, passivation, and deposition all occur
separately.6 Existing procedures have exhibited the ability of alkanethiols to react readily
with reactive sites on Ge(111) and Ge(100), producing stable monolayers.6, 65, 66, 73
2.3 Experimental Methods
Recent studies have shown that the deposition of alkanethiols on germanium is a
reproducible process resulting in monolayers on technologically relevant substrates.
Furthermore, the interaction between thiol solution and Ge(100) has been shown to
construct a monolayer with superior stability and thickness to those on Ge(111).6
Through control of reaction conditions, we have improved upon pre-established methods
for alkanethiol self-assembly on Ge(100) and succeeded in simplifying the procedure.
25
2.3.1 Materials
Germanium(100) samples were purchased from Silicon Quest International (Santa
Clara, CA). They were 350-400 μm thick and had resistivities of 40 Ω-cm. Absolute
pure ethanol and 1-dodecanethiol (C12) were acquired from Sigma Aldrich (St. Louis,
MO). Deionized water at 18.2 MΩ-cm resistivity was provided by a Milli-Q system from
the Millipore Corporation (Bellerica, MA).
2.3.2 Fabrication of Monolayers
Solutions are made via a two-step process for all measurements. A 1 mM
ethanolic solution of C12 is prepared, which is then diluted with 18.2 MΩ-cm water,
resulting in a 0.5 mM C12 solution. The ethanol and water ratio is 1:1, unless otherwise
specified. In cases where the ratio was varied, a measured quantity of C12 was dissolved
in ethanol, and then diluted with water to the specified volume, while C12 concentration
was kept at 0.5 mM. Adding C12 to a pre-prepared ethanol/water solution can produce
monolayers in many cases, but as C12 becomes less soluble in the mixture, phase
separation decreases reproducibility. Our procedure is preferable, because it ensures the
dissolution of C12 over a wide range of solvent ratios.
Samples of germanium, on the order of 9 mm2 in area, were cleaved from a
Ge(100) wafer along its crystallographic planes. The sample size was determined such
that three separate contact drops could be placed on the surface without encountering a
previously tested region. The samples were placed in a vial with the ethanolic solution
and left alone for a pre-determined amount of time. A 24 hour deposition time can be
26
assumed unless otherwise specified. Samples were removed from the solution, rinsed
with ethanol, blown dry with nitrogen gas, and placed in a clean vial until needed.
2.3.3 Volcano, Kinetics, and Stability Measurements
To decipher the role water plays in the deposition process, we ran a series of
measurements with varying ratios between water and ethanol. Solutions of C12 in the
solvent/water mixtures were created for all desired ratios. Germanium samples were
placed into each solution for 24 hours, and were analyzed by contact angle measurement.
To analyze the timescale of self-assembly between C12 and Ge(100), we
measured the contact angle for samples experiencing varying deposition times. A
germanium sample is placed in a 1:1 solution of water/ethanol overnight to remove its
oxide layer. It is rinsed, dried, and placed into a 0.5 mM solution of C12 in a 1:1 mixture
of water/ethanol for a time interval. The sample is removed, rinsed, dried, and analyzed.
It is placed back into solution until the next time interval is reached. This procedure is
repeated at specific time intervals for the remainder of the experiment.
Alkanethiolate SAMs will degrade over time if exposed to ambient conditions. It
was suggested by Bent and co-workers that exposure to light accelerates the degradation
process.6 We tested this claim by monitoring contact angle stability on identically
prepared samples, protecting some from light exposure while exposing others to light. An
ethanolic solution of C12 was prepared with a 1:1 ratio of water/ethanol and a 0.5 mM
concentration. Germanium samples, cleaved from a single wafer, were placed into the
solution for 24 hours, before being removed, rinsed, and blown dry. Initial contact angle
measurements were taken. Both samples were placed in transparent glass containers; one
27
was left as is, while the other was covered in foil. Every 24 hours, the samples were
removed, rinsed with ethanol, blown dry, evaluated through contact angle analysis, and
then placed back into their respective vials.
2.3.4 Contact Angle Measurements
Our measurements were collected using a custom-built goniometer, encompassing
a 0.50× magnification, 94 mm focal length InfiniStix (Hitachi Ltd., Tokyo, Japan) CCD
camera. A flat-tip syringe-needle (33 gauge) is attached to a .2 mL Gilmont micrometer
syringe (Cole-Parmer, Vernon Hills, IL) by means of Torr Seal (Varian, Inc., Palo Alto,
CA) epoxy. Data were acquired by National Instruments (Austin, TX) Measurement and
Automation software. Angles were measured using ImageJ software from Scion Image
(Frederick, MD).
We chose to report advancing and receding angles separately, rather than
reporting an equilibrium angle. The first step in our procedure was to deposit a 5 μL
droplet of deionized water onto our sample surface. The needle was re-positioned to the
center of the drop carefully, as to not disrupt its shape. The drop’s volume was increased
to 10 μL. An image of the droplet was recorded. Next, the droplet was expanded until the
wetted area of the surface increased. The value for which this typically occurred was
between 15-20 μL. Again, the needle was centered in the drop. The volume was reduced
to 10 μL, and again photographed. This procedure was repeated for three separate
locations on the sample surface to account for any physical or chemical heterogeneity.
The initial image was analyzed as the advancing angle, and the latter image as the
receding angle. Contact angles from both the left and right sides of the droplet were
28
measured and averaged together, resulting in that angle’s value. The averages of the three
advancing and receding angles were averaged and resulted in one advancing and one
receding value for each substrate. Figure 2.2 illustrates the measurement procedure
incrementally.
2.3.5 X-Ray Photoelectron Spectroscopy Analysis
Samples submitted for XPS analysis were kept in solution until being loaded into
the XPS chamber. The base pressure of the chamber was set at 1 × 10-9
Torr. The
samples were analyzed by a Kratos Axis Ultra photoelectron spectrometer using a
monochromatic Al Kα (20 Ma, 14 kV) source and using a 300 μm × 700 μm spot size.
Survey spectra were analyzed using a pass energy of 80eV (1 eV resolution) and
high-resolution spectra of the Ge 3d, Ge 2p, C 1s, O 1s, and S 2p by a pass energy of 20
eV (0.1 eV resolution). The spectra peaks were fit using Gaussian-Lorentzian (GL) line
shapes and a Shirley background in CasaXPS analysis software.60, 74
2.4 Results and Discussion
Self-assembly on germanium can occur in ambient conditions only in the absence
of its native oxide layer. To accomplish this, previous work has demonstrated the success
of a multi-step process. The procedure includes separate phases for removing oxide
layers, passivating the surface against re-oxidation, and the deposition of alkanethiols.
The oxide layer is water soluble and is removed by rinsing the surface with H2O.6
Passivation of germanium is accomplished by deposition of halogen atoms, as well as by
treatment with chlorine or bromine, by means of hydrochloric or hydrobromic acid,
29
Figure 2.2: Step-by-step method of measuring contact angles by the dynamic sessile
drop method. A. Syringe is located over an un-tested region. B. A 5 μL drop is deposited
onto the substrate surface. C. The syringe is carefully centered within the drop. D. The
drop volume is increased to 10 μL, and the advancing angle measurement is taken. E.
The drop volume is increased to 15 μL (or until the wetted surface in increased). F. The
drop volume is decreased to 10 μL; the receding angle is measured.47
A
Position Sample
F
Decrease to 10 μL
E
Increase to 15 μL
B
Deposit 5 μL
D
Increase to 10 μL
C
Center Needle
θA θR
30
respectively.6, 65, 73
Hydrogen-terminated surfaces are formed from treatment with
hydrofluoric acid. Passivating the surface slows down the re-oxidation process, allowing
sufficient time for thiol deposition, before oxide reformation. It is critical to note that
SAM deposition is dependent upon the absence of its native oxide, and not the presence
of a passivating layer.
We introduce a process where removal of the oxide layer and self-assembly occur
simultaneously, in the absence of a pre-formed passivation layer. The native oxide layer
on germanium and the alkanethiol adsorbate have conflicting solubilities. Germanium
oxide is soluble in water, but not in ethyl alcohol. Alkanethiols fail to produce
homogenous mixtures in water, but will do so if mixed in ethyl alcohol. By mixing water
and ethanol, we produce an environment for which both the adsorbate and oxide layer are
soluble. This solution can dissolve the oxide layer, and enable deposition in a single step.
Alkanethiol SAM formation cannot occur successfully if the oxide layer fails to be
completely removed, or if the adsorbate is insoluble. Therefore, alcohol and water are
both necessary constituents in the thiol solution. Figure 2.3 shows our method for C12
SAM deposition on germanium.
2.4.1 Mixed-Solvent System
We designed an experiment that would yield insight into both the ability of self-
assembly to occur in a mixed-solvent environment, and which water/alcohol ratios would
satisfy both oxide and thiol solubility. We analyzed mixtures between water and three
solvents; ethanol, methanol and t-butanol. Figure 2.4 shows “volcano plots,” expressing
the relationship between solvent/water ratio and the SAM quality. The t-butanol data is
31
Figure 2.3: The importance of solubility on direct SAM deposition on Ge(100).
Germanium oxide is soluble in water, while the alkanethiols are soluble in alcohols. A
water/alcohol mixture can solvate both the germanium oxide and enough alkanethiol to
form full-coverage SAMs. The unmixed solvents (and exposure in a stepwise fashion) do
not produce SAMs.25
32
Figure 2.4: Dependence of SAM quality on mixed solvent ratio. The appropriate
quantity of C12 is dissolved in the ethanol fraction, and then diluted with water to the
selected volume and ratio. Methanol yields a wider window of solvent composition than
ethanol. Films with an advancing contact angle greater than 100° are considered to be of
high quality. t-Butanol results are not shown, as no ratio resulted in complete films.25
33
not shown as it failed to produce any complete films. Advancing contact angles equal to
or greater than 100° are considered to be indicative of good monolayer coverage, and a
quality SAM.75, 76
Quality monolayers were formed for a methanolic solution containing
20-80% water. Ethanol only produced good SAMs for water percentages between 40%
and 60%. Methanolic solutions exhibited the ability to out produce ethanolic
counterparts for both the case of low and high water concentrations. We propose that in
low water/high alcohol situations, germanium oxides are more soluble in mixtures
containing methanol rather than ethanol or t-butanol. Higher solubility allows complete
removal of the native oxide layer and drives self-assembly. Intuition would suggest that
for high water mixtures, self-assembly would improve for an increase in alcohol
hydrophobicity, due to the hydrophobic alcohol’s ability to keep alkanethiols solvated.
However, the data showed that the more hydrophobic alcohols, ethanol and t-butanol,
failed to form SAMs under these conditions. We believe that these results suggest a
deposition process dependent upon more than simple solubility, and further study will be
required to determine the mechanism.
2.4.2 Deposition Kinetics
The coverage and quality of monolayer films can be tracked by contact angle
measurements, enabling the study of the relationship between the growth of a SAM and
its solution exposure time. It is accepted that full-coverage n-alkanethiol SAMs produce
contact angles of ≥100°.75, 76
Contact angle measurements on bare, as-received
germanium are affected by the instability, and presence of a native oxide layer. In the
case of virgin contact between water and germanium, contact angles are indicative of the
34
hydrophobic nature of GeO and GeO2. Additional drops, as a result of oxide removal,
will begin to exhibit the strongly hydrophilic characteristics of a pure Ge(100) surface.
This phenomenon is shown in Figure 2.1. Short solution exposure times were inadequate
for removal of the oxide layer. Before kinetics measurements were taken, to ensure that
the pure germanium surface was being studied, the sample was introduced to a 1:1
water/ethanol solution for 24 hours. By stripping the oxide layer, any increase in
hydrophobicity can be attributed to the deposition of C12. Figure 2.5 shows the growth of
a monolayer on germanium with time. A period of rapid growth occurs between 10 and
1000 minutes, with little to no improvement in monolayer quality after 24 hours. High
coverage films are observed after 6-18 hours. Self-assembly on germanium is a slow
process in comparison to gold, where alkanethiolates achieve full coverage within
seconds.12
The time for full coverage to occur varied between runs, but complete
monolayers were achieved within 24 hours for all cases.
2.4.3 Long Term Stability
Self-assembled monolayers will degrade over time if not protected from ambient
conditions. Our one-pot synthesis method for depositing SAMs on germanium produced
monolayers with stability characteristics comparable to those reported in previous work,
leading us to suspect that the degradation process will also be similar. Bent and
co-workers suggest that defects or impurities in the monolayer allow ambient oxygen to
interact with the germanium surface, and cause re-oxidation.6 Our measurements are
consistent with this suggestion. Once the oxygen accesses the substrate and initiates
re-oxidation, the SAM loses its ability to protect the surface. Degradation occurs until the
35
Figure 2.5: Kinetics of SAM formation on Ge(100). Deposition is from 0.5 mM C12
solution in 1:1 ethanol/water. Advancing water contact angles are shown on the
logarithmic scale for clarity. The period of most dramatic change occurs between 10 and
1000 min., and run-to-run variation occurs in this region.25
36
entire surface is once again covered in an oxide film similar to that of the as-received
sample. Bent also suggested that exposure to light caused the monolayer to degrade more
rapidly.6 To test this claim, we exposed samples of identical preparation parameters to
ambient conditions. One sample was exposed to light, and the other confined to
darkness. Figure 2.6 shows our data for the degradation of monolayers in relation to
varying light conditions. Degradation is present in both cases, but no significant
differences in the rate of re-oxidation are apparent. This leads us to believe that mild light
exposure has little bearing on the process, and that defects in the monolayer are the
dominant factor for re-oxidation rate.
2.4.4 X-Ray Photoelectron Spectroscopy Analysis
X-ray photoelectron spectroscopy analysis is used to confirm the formation of
SAMs on germanium and the presence/removal of germanium oxide. Figure 2.7 shows
the high-resolution spectra of the C 1s, S 2p, Ge 2p, Ge 3d, and O 1s orbitals. These
spectra are intrinsic to the elements we wish to identify. Carbon and sulfur are indicative
of the C12 monolayer. Germanium and oxygen provide substrate information. Each
individual spectra is overlaid with three peaks representing the as-received germanium
sample (black trace), the oxide-free, bare Ge sample (blue trace) produced from rinsing in
1:1 water/ethanol solution, and SAM sample (red trace).
The C 1s and S 2p spectra will be discussed first. As expected, the as-received
germanium sample exhibits a broad carbon peak. The majority of adventitious carbon is
adsorbed on the germanium oxide. Removing the oxide layer also acts to rid the surface
37
Figure 2.6: Advancing contact angles collected after days of exposing C12 SAMs on
Ge(100) to the ambient environment, with samples either exposed to or protected from
light. There was no significant difference between the two series.25
38
Figure 2.7: High-resolution XPS spectra collected for the regions specific to the C 1s,
S 2p, Ge 3d, O 1s, and Ge 2p. Each region is shown with its own arbitrary intensity scale.
A. Adventitious carbon (black trace) is effectively eliminated by treatment with the
water/ethanol (blue trace), enabling a high-purity C12 SAM to form (red trace). B. The
C12 monolayer gives the expected response in the S 2p region, while the as-received and
the cleaned wafers do not show the presence of any sulfur. C. Oxidized species (red
trace) are removed readily by the water/ethanol treatment, leaving high intensity peaks
attributed to the bulk germanium crystal (blue trace). The intensity is attenuated by the
presence of the SAM (red trace). D. The Ge 3d spectra provide information
complimentary to that of the Ge 2p spectra. Water treatment again eliminates the signal
from the oxidized Ge species (blue trace). The Ge 3d doublet appears shifted slightly to a
lower binding energy. The intensity is again attenuated by the presence of the monolayer
(red trace). E. The majority of the oxygen is attributed to the native germanium oxide;
after removal or SAM formation, remaining oxygen is attributed to residual alcohols or
oxide reformation at defect sites before the samples were loaded into the XPS vacuum
chamber.25
39
of carbon contamination, evident by the absence of a strong carbon signal for the bare Ge
sample. The SAM sample exhibits a sharp peak at 284.9 eV, typical of carbon atoms in a
C12 monolayer.77, 78
The direct Ge-S bond formed by alkanethiol self-assembly on
germanium is represented by the S 2p3/2 and S 2p1/2 peaks, at 162.4 and 163.6 eV,
respectively. The presence of the aforementioned sulfur peak, and the lack of sulfur
signatures for the other germanium samples is in agreement with spectra measured for
thiolates on gold.75, 78-80
The Ge 2p, Ge 3d, and O 1s peaks inherent to germanium and its oxide will now
be discussed. The variation in the Ge 2p3/2 and Ge 2p1/2 peak intensities are indicative of
the progression of the self-assembly formation procedure. Both bulk germanium (1218.0
and 1249.0 eV) and oxidized germanium (1220.9 and 1252.0 eV) are present for the
as-received sample.63
The cleaned sample exhibits a strong reduction in the germanium
oxide peaks and a growth in bulk germanium peaks. The decrease in the oxide peak
suggests that the solvent/water mixture successfully removed the native oxide layer. The
removal of the oxide layer provides photoelectrons generated in the bulk germanium an
un-inhibited pathway to the detector, explaining the dramatic increased peak intensity.
The remainder of a germanium oxide peak is attributed to the brief ambient exposure the
sample receives before loading in the vacuum chamber. The SAM sample again
experiences a decrease in the bulk germanium peak intensity due to the thickness of the
C12 monolayer, blocking some photoelectrons from the bulk germanium. The
degradation and short life times of SAMs on germanium can be attributed to the gradual
growth of germanium oxide in defects of the SAM after assembly, shown to be present in
the form of a weak shoulder at 1220.9 eV.
40
The Ge 3d region expresses similar characteristics to those of the 2p peak. The
Ge 3d peak also experiences splitting in the form of the Ge 3d5/2 (29.95 eV) and Ge 3d3/2
(30.53 eV) doublet, but the peaks are not well-resolved.63, 81, 82
The oxide layer appears in
the form of GeO2 with a peak at 33 eV in the as-received sample. The presence of oxide
is negligible for both the cleaned surface and post-deposition samples. The Ge 3d peak
experiences a slight shift to higher binding energy after the removal of its oxide layer, a
result of passivation layer elimination. Again, the bulk germanium peak experiences a
reduction in intensity after deposition occurs.
Finally, the O 1s peak exhibits a strong intensity at 532.5 eV for the as-received
sample. Its intensity decreased, but some oxygen is still present after oxide removal and
SAM deposition. This is attributed to reformation of oxide due to ambient exposure, to
the presence of defect sites in the SAM, and to residual adsorption of ethanol.
2.5 Conclusions
We have demonstrated the ability to deposit well-formed alkanethiol monolayers
on Ge(100) surfaces by a single-step deposition technique. The critical factor for
successful deposition is the adjustment of the water/solvent mixture such that both the
oxide layer and adsorbate species can be efficiently dissolved. Our ultimate goal is to
produce a self-assembly technique for germanium that produces monolayers mimicking
the advantages of those of thiols on gold. Alkanethiols on Au{111} present a model
system for self-assembly for several reasons, particularly the ease with which quality
films can be deposited. We propose that simple self-assembly can be extended to a
number of substrates if the proper chemistries are developed.
41
Chapter 3
Chemical Patterning on Germanium(100)
3.1 Overview
The recently realized potential of micro- and nano-arrays on solid substrates has
driven the trend of electronic miniaturization.83, 84
Self-assembly is a powerful approach
for nano-scale device construction. Recently, self-assembled monolayers have attracted
increasing interest for their ability to interact with, and to control, interfacial chemistry of
semiconductor substrates. The current challenge is the preparation and integration of thin
films into semiconductor-based devices. Chemical patterning techniques, especially
μCP, have been used to produce single-molecule-thick nanostructures with micron (or
better) resolution on technological substrates, exploiting both the stable nature of SAMs
and the ability to place individual molecules precisely.50, 54, 85, 86
Patterning processes
enable the fabrication of arrays and the manipulation of interfacial properties.
3.2 Previous Work: Chemical Patterning and Thin Film Deposition
Alkanethiols on gold are widely recognized as the model system for
self-assembly. It follows that the majority of chemical patterning and film deposition
techniques have been engineered to facilitate this chemistry. With the development of
the integrated circuit, an ever-increasing need for solar energy, and a trend toward the
miniaturization of electronic components, novel techniques for patterning on
semiconductor substrates are desirable. Recently, research has begun to manipulate
pattering techniques originally intended for thiols on gold, extending them to other
materials. This emerging field has great growth potential. Some of the more prevalent
42
techniques for patterning on semiconductor substrates discussed in recent literature are
worth examining to set the current work in perspective.
3.2.1 Vapor Deposition and Etching
Thin films can be used to stabilize semiconductor surface sites and to improve
electrical properties, including carrier mobility and conductivity. Deposition of thin films
is the most direct way to obtain high-κ dielectric layers on both silicon and germanium.87
Not only are the creation of direct substrate/dielectric interfaces possible, but such thin
films may act as the foundation for etching and lithography methods.
Vapor deposition is a frequently implemented technique for thin-film deposition.
Vapor deposition techniques are categorized under two broad terms; chemical vapor
deposition (CVD),87-93
and physical vapor deposition (PVD).90
The former depends on
surface chemical reactions to deposit material, and the latter employs physical methods
such as sputtering to construct films.90
Chemical vapor deposition occurs in several
forms; plasma-enhanced chemical vapor machining (CVM),93
ultrahigh vacuum CVD
(UHV-CVD),92
low-pressure CVD (LPCVD),91
and metal-organic CVD (MOCVD),87
amongst others. Physical vapor deposition can also be found in many forms, including
pulsed-laser deposition (PLD)94
and sputter deposition.90
The ability to insert certain materials into electronic devices is first dependent
upon thin-film deposition, with patterning a secondary concern. Vapor deposition is not a
direct patterning technique, but it offers a method for growth of dielectric films on
semiconductor substrates. The resultant films lend themselves to patterning procedures.
Etching is a patterning technique often associated with thin films. It utilizes patterning
43
masks and chemical reactions to remove surface layers selectively.92
Zanier and
co-workers demonstrated the utility of electron-beam lithography for the patterning of
square lattices of circular dots onto a photoresist. Resist dots were used as a master to
transfer the pattern onto a SiGe substrate. Reactive ion etching (RIE) uses the gas
mixture SF6/CHF3 to remove the top layers of the unmasked SiGe surface, leaving a
periodic lattice of dots on the substrate.92
Such RIE procedures are critical methods for
the subtractive patterning of surfaces.91-93, 95-97
Etching is also used to remove
undesirable surface layers, so that functional adsorbates can be deposited.6, 98-101
The
mask designs and etching chemicals are varied in accordance with preferred patterns and
materials. Etching meets requirements for both the ability to pattern on semiconductor
substrates and production of nanometer features.93, 95
Similar methodology will be
revisited later in section 3.4, where we selectively remove germanium’s native oxide
prior to chemical patterning.
3.2.2 Additional Patterning Techniques for Semiconductors
A number of soft-lithography techniques have been used to create controllable
microstructures, though feature sizes below 100 nm remains an important challenge.86
Dip-pen nanolithography (DPN) has been used to fabricate nanoscale features,
and has recently moved beyond gold substrates and onto semiconducting and insulating
materials.85, 86, 102, 103
Dip-pen nanolithography utilizes an ink-soaked atomic force
microscope (AFM) tip to direct-write (and thus to pattern) molecular adsorbates onto
surfaces. Physical or chemical reactions between the ink-soaked tip and substrate cause
materials to immobilize and to bind to the surface. Because the reaction occurs at the
44
AFM tip, material adsorption will localize along the tip’s trace pattern.86
In the case of
standard DPN (as opposed to newly developed techniques, including electrochemical
DPN), the ink is transported from the tip to the surface through a medium, typically a
water meniscus.85, 86, 103
This technique is excellent for the production of
serially-produced features with sub-100 nm dimensions, even depositing line widths as
small as ~10 nm.85
Wang and co-workers demonstrated the ability to grow and to pattern germanium
nanowires on various substrates via a gold-seeded vapor liquid solid (VLS) technique.88,
89 CVD was applied, using germane (GeH4) gas, forming an alloy through the dissolution
of germane into a Au cluster. Further exposure causes the alloy to liquefy, followed
rapidly by the supersaturation and precipitation of the germanium, resulting in the axial
growth of single-crystal nanowires.88
They proposed two techniques for patterning the
germanium nanowires onto substrates. First, e-beam lithography is used to pattern small
arrays of Au atoms into islands. The islands will nucleate into ordered dots upon
annealing. The germanium nanowire growth process can be performed on the Au dots.88
A second patterning technique involves germanium nanowire passivation and the
use of Langmuir-Blodgett troughs to form ordered assemblies of the wires. Again, CVD
was used to grow nanowires on silicon substrates. The nanowires were removed and
cleansed via ultrasonification and etching, respectively.89
The wires are passivated by
alkanethiols through the use of a Grignard reaction. The passivated nanowires are placed
into a Langmuir-Blodgett trough filled with an ethanol/water mixture, where they form a
close-packed, dense film oriented parallel to the trough edges. From this state, the wires
45
can be transferred to many substrates.89
Even if a non-semiconducting substrate is used,
the sample will retain the semiconducting character of the germanium nanowires.
Passivated germanium nanowires have exhibited the ability to endure patterning, offering
an alternative route to device fabrication.
3.2.3 Microcontact Printing on Germanium
Microcontact printing is the technique utilized for patterning on germanium in
this dissertation. As described earlier, it has potential for a wide range of microelectronic
applications. We chose μCP for several reasons. Although traditional lithographic
techniques have supplanted it for maximum resolution, it remains the pre-eminent
method for producing well-ordered and stable chemical patterns with microscale features
over long ranges. Secondly, μCP lends itself to the technique we devised for
self-assembly on germanium in chapter 2. Finally, despite the large body of work for
alkanethiol deposition on gold, there is comparatively little work on μCP of molecular
inks on semiconductor substrates.104
Wang and co-workers successfully deposited nanometer-scale patterns of SAMs
onto amorphous Si, crystalline Si, and SiO2 substrates, by means of μCP.104
Their
simple methodology employed a patterned polydimethylsiloxane elastomeric stamp, a
octadecyltrichlorosilane (OTS) molecular ink, and the use of potassium hydroxide for
etching. They provide a demonstration confirming that μCP on semiconductor surfaces is
possible. In each of these cases, silane reacts with the native silicon oxide, a robust oxide.
Patterning on germanium, however, is complicated by its native oxide layer.
Recall from the previous chapter that the water soluble and unstable germanium oxide
46
precludes the use of strategies akin to those employed on silicon. As a result, there are no
reports of direct chemical patterning of molecular inks on germanium by μCP in the
literature. Modifications to the microcontact printing procedure have been developed,
however. Porter and co-workers demonstrated a method for negative patterning of thin
Au, Pd, and Pt nanoparticle films on Ge(100), by μCP.83
They present a negative μCP
technique in which an aqueous solution of metal salt is placed onto a germanium surface
and patterned by a PDMS stamp. A solution of PdCl42-
was dropped onto the germanium
surface and immediately contacted by an oxidized, hydrophilic PDMS stamp.83
The salts
undergo slow re-oxidation, resulting in spatially defined features of metal particles,
located in non-contact regions.
3.3 Experimental Methods
There exists a disproportionate relationship between the importance of patterning
on germanium and the amount of work devoted to studying it. Application of soft
lithography techniques has significant potential for the development of new devices. By
manipulating standard μCP procedures, we have developed a method that enables us both
to remove germanium’s native oxide layer and to deposit alkanethiol SAMs,
simultaneously.
3.3.1 Materials
Germanium(100) and silicon(100) wafers were purchased from Silicon Quest
International (Santa Clara, CA). They were 350-400 μm thick and had resistivities of 40
Ω-cm. Absolute ethanol, 11-mercaptoundecanoic acid (MUDA), and 1-dodecanethiol
47
were acquired from Sigma Aldrich (St. Louis, MO). Deionized water at 18.2 MΩ-cm
resistivity was provided by a Milli-Q system from the Millipore Corporation (Bellerica,
MA). Hydrochloric acid of 36.46 molar mass was acquired from EMD Chemicals, Inc.
(Gibbstown, NJ). Polydimethylsiloxane stamps were produced in-house, using PDMS
Sylgard 184 base and PDMS Sylgard 184 cure (Dow Corning, Midland, MI).
3.3.2 Stamp Preparation
The PDMS polymer was made in accordance with procedures described
elsewhere.105
Stamps were patterned using Si/SiO2 masters. Micron-scale relief features
on the master were generated using photolithography and reactive-ion etching. Before
stamp production, the silicon master was covered with trichlorosilane to inhibit PDMS
adhesion.105
The polymer was poured onto the master, and subsequently degassed in a
vacuum until all air bubbles were removed. The PDMS was placed in an oven, set to
90°C, for 2 hours.
A 25 mM ethanolic solution of C12 was prepared. PDMS stamps were placed in
the solution overnight, allowing alkanethiols to diffuse into the stamp. The PDMS
stamps were removed from the solution, rinsed and blown dry. The stamps were placed
into a beaker containing a 1:1 water/ethanol solution. They were arranged such that the
patterned faces were fully exposed to the solution. The stamps were ultrasonicated for
three minutes to remove surface contamination. They were removed from the sonicator,
rinsed, and blown dry.
Just before use, the stamps were further cleaned by a blotting technique. A silicon
wafer was treated in an ultraviolet (UV)-ozone oxidation cleaner (UV clean model
48
135500, Boekel Scientific, Feasterville, PA) for ten minutes to prepare the clean silicon
surface for blotting. The stamp was blotted by mechanically pressing the patterned face
against a virgin region of the silicon wafer. After careful removal (making sure not to
touch the blotted face) from the silicon surface, the stamp was ready for printing.
3.3.3 Sample Preparation
Samples of germanium were cleaved from a Ge(100) wafer along its
crystallographic planes. The samples were placed in a 1:1 water/ethanol solution
overnight. They were removed, rinsed with neat ethanol, blown dry with nitrogen, and
placed in a petri dish, making sure the polished germanium surface was facing upward.
For submerged μCP (SμCP), water at 18.2 MΩ-cm resistivity was added to the petri dish
so as to submerge the sample completely. The sample remained undisturbed for at least
10 minutes.
3.3.4 Stamping Procedure
The PDMS stamp was placed on top of the prepared sample such that the
patterned side of the stamp and the polished face of the sample were in contact. A weight
was placed on top of the stamp to ensure conformal contact between it and the sample for
the duration of the experiment. When the desired stamping time interval was met, the
samples were removed from the petri dish, peeled from the stamp, rinsed with ethanol,
blown dry, and placed in a clean vial until analysis.
For backfilling, samples prepared at 24 hour stamping intervals were submerged
in a 0.5 mM 1:1 ethanol/water solution of MUDA for 24 hours. The sample were
49
removed from solution, rinsed, blown dry, and placed into a clean container, until
scanning electron microscopy (SEM) analysis.
3.3.5 Analysis Techniques (XPS, SEM)
X-ray photoelectron spectroscopy measurements were performed to determine the
coverage of μCP SAMs on germanium. Samples submitted for XPS analysis were
produced from time-dependent experiments, described below. Because these samples
could not be replaced into solution after termination of stamping, the procedure was
carefully formulated to minimize the time between removal from solution and XPS
analysis. The order of the experiments was arranged such that sample requiring
maximum stamping time was initiated first, followed by the next longest exposure, such
that all runs would conclude at approximately the same time. The XPS analysis
procedure is identical to the measurements performed in chapter 2. For this experiment,
only the C 1s region is reported, as the results from the other regions are similar to what
was already shown.
Scanning electron microscope images were used to determine the stamping
mechanism’s ability to transfer pattern features. Field-emission scanning electron
microscopy (FESEM) images were collected using a LEO 1530 Gemini System (Carl
Zeiss, Inc., Oberkochen, Germany). The system was maintained at an operating voltage
of 5 kV. Primary electron-beam currents of 125 pA at 1 kV were used to collect images,
utilizing an in-lens secondary electron detector. Sample exposure to the e-beam was
limited to no more than 20 s to limit sample damage. The electron detector was
maintained at a collection voltage of +300 V, with an aperture size of 30 μm.
50
3.4 Results and Discussion
Microcontact printing has the ability to produce well-ordered and stable SAM
patterns of microscale features50
, but little work has been done to test its applicability on
germanium. The lack of patterning work on germanium can be attributed to an intrinsic
oxide layer that resists pattern deposition. In chapter 2, we detailed a method for
successful formation of monolayers through concurrent alkanethiol deposition and oxide
layer removal. Manipulation of μCP, in the form of SμCP, enabled us to reproduce those
results, demonstrating a μCP technique directly applicable to germanium.
3.4.1 Technique Progression
To test the ability of μCP to transfer pattern features and produce full-coverage
monolayers on germanium, we analyzed samples with XPS and SEM. Because XPS is an
ensemble technique, unpatterned (flat) stamps were used for monolayer deposition, so as
to create a uniform surface. First, we performed μCP on as-received germanium. This
method failed to produce a monolayer. A second sample was introduced to a water bath,
and stamped in the presence of the water medium, described schematically in Figure 3.1.
This submerged microcontact printing technique was utilized so as to remove the native
oxide layer and inhibit reoxidation.106
We attempted to improve the technique by
submerging the germanium in hydrochloric acid (HCl).
51
Figure 3.1: Schematic comparing standard microcontact printing (a) and submerged
microcontact printing (b). A. PDMS stamp is produced through introduction to an etched
silicon wafer. B. The stamp is saturated by an ethanolic C12 solution. C. In standard
microcontact printing, the stamp is mechanically pressed against the substrate under
ambient conditions. Our method involved contact between stamp and germanium in the
presence of water, so as to eliminate the native oxide. D. Adsorbate pattern is deposited
onto substrate in atmospheric air (a), and while submerged in water (b).106
A
B B
C
D
52
3.4.2 X-Ray Photoelectron Spectroscopy Analysis
As was addressed in chapter 2, the photoelectrons generated by the carbon atoms
of alkyl chains exhibit a binding energy of 284.9 eV. The intensity of that peak
corresponds to the population of the C12 alkyl chains, and provides insight into
monolayer coverage. Figure 3.2 shows the high-resolution spectra of the C 1s orbital,
observed for samples produced by submerged microcontact printing in water. The figure
is inlaid with five spectra; the as-received germanium sample (red trace), the 1:1
water/ethanol rinsed bare germanium sample (green trace), full coverage SAM on
germanium (purple trace), no-ink stamp sample (control, blue trace), and microcontact
sample (black trace).
Previously addressed in chapter 2 was the adventitious adsorption of carbon by
the native oxide layer. Again, we see a significant carbon signature for the as-received
sample. After rinsing, the oxide is removed, along with its adsorbed carbon, explaining
the strong decrease in peak intensity for the bare germanium sample. The SAM sample
shows the highest intensity of any peak, evident of its high-coverage monolayer. The
SμCP sample exhibits a peak approximately one-half the intensity of the fully formed
monolayer and positioned at an identical energy value. The correspondence to the C12
sample, in terms of location, indicates similarity between carbon atoms, suggesting the
presence of a partial monolayer. The lower peak intensity shows that the monolayer
achieved only 50% coverage. A control sample, patterned with an un-inked PDMS
stamp, confirms that the inking solution was the sole source of carbon deposition. The
control sample shows a minimal peak, suggesting limited influence on deposition.
53
Figure 3.2: High-resolution XPS spectra collected for the region specific to the C 1s
orbital. The adventitious carbon of the as-received sample (red trace), is effectively
removed by rinsing with water/ethanol (green trace), allowing a high-quality C12
monolayer (purple trace) to form. Our method for submerged microcontact printing on
germanium produces a monolayer of 50% coverage (black trace).64
1_Ge
_recd [0] 2_Ge_EtOH_
H2O [0] 3_Ge_C12_S
AM [0] Ge_uCP_co
nt_a [1] Ge_uCP_co
nt_b [1] Ge_uCP_C
12_a [1] Ge_uCP_C
12_b [1]
2
90 2
80 Binding Energy (eV)
C 1s
As Received
Oxide Removed
C12 Monolayer
Control
Printed Monolayer
54
In an attempt to improve monolayer coverage, the sample was submerged in HCl for
SμCP. Attempting to chlorinate the surface did not improve the SAM coverage.
3.4.3 Scanning Electron Microscopy Analysis
Scanning electron microscopy images were used to analyze patterned germanium.
Figure 3.3 shows a FESEM image of a C12-patterned germanium substrate. The SEM
images show square patterns exhibiting crisp, clean edge fidelity, indicative of a nearly
perfect transfer of stamp features.
Despite the excellent pattern transfer, features were still at submonolayer
coverage. Backfilling stamped samples with MUDA was performed to heal empty
locations, both inside and outside regions of stamp contact, and thus to protect the
integrity of the sample surface. The resulting SEM image, shown in Figure 3.4, showed
low-intensity squares on a higher intensity background (an inversion of Figure 3.3). The
inversion of pattern color suggests MUDA monolayer formation.50
The moderate
contrast between patterned and filled locations results from the low surface coverage of
the patterned SAMs. Through backfilling, we showed the ability to control the chemistry
both inside and outside of the square pattern features. Furthermore, the retained pattern
fidelity is indicative of C12 monolayer stability, and the gentle and non-damaging nature
of the backfilling method.
55
Figure 3.3: Scanning electron microscopy images of C12 monolayers, patterned on
germanium (100). Images represent a single patterned surface at varying resolutions.64
56
Figure 3.4: Scanning electron microscopy image of a C12 monolayer, patterned on
germanium (100), backfilled with MUDA. The inversion of intensity, in comparison
with the untreated C12 monolayer, is indicative of MUDA SAM formation. Successful
backfilling represents our ability to control the chemistry of both the patterned and
non-patterned regions. Retained edge fidelity of the patterned squares shows the
non-damaging nature of the backfilling procedure and the stability of the monolayers. A
lack of C12 monolayer coverage results in limited contrast between regions.64
57
Figure 3.5: Scanning electron microscopy image of a C12 monolayer, patterned on
germanium (100), backfilled with MUDA. This image is a contrast enhanced version of
Figure 3.4, so as to simplify feature decipherability. The inversion of intensity, in
comparison with the untreated C12 monolayer, is indicative of MUDA SAM formation.
Successful backfilling represents our ability to control the chemistry of both the patterned
and non-patterned regions. Retained edge fidelity of the patterned squares shows the
non-damaging nature of the backfilling procedure and the stability of the monolayers. A
lack of C12 monolayer coverage results in limited contrast between regions.64
58
3.5 Conclusions
We have successfully demonstrated the ability to pattern alkanethiol monolayers
on Ge(100) through utilization of submerged microcontact printing. Transferring
patterns at 100% coverage remains a priority. Further investigation is required to identify
the factors inhibiting coverage. Re-oxidation of the germanium surface has been
suggested as one possible cause for a lack in coverage. The failure to produce films on
bare germanium is consistent with this hypothesis, as the substrate is not protected from
oxygen. In SμCP, re-oxidation may occur as water is expelled at contact sites, and thus
can no longer remove oxide in these areas. Upon contact, the stamp may displace the
water medium, allowing contact between surface and oxygen. This was expected to have
been resolved by the passivating ability of HCl, but no improvements in coverage were
observed. Preliminary work has indicated that a correlation between stamp and surface
wettability may improve results. With continued work, we hope to produce patterned
assemblies comparable in order, coverage and functionality, to both those of thiols on
gold, and the ones we deposited on germanium.
59
Chapter 4
Conclusions and Future Prospects
4.1 Review
Techniques for self-assembly and chemical patterning are largely oriented toward
the deposition of alkanethiols on gold substrates because of the well-ordered, stable, and
chemically functional monolayers produced at the thiol/Au interface. The ability to
deposit and to pattern stable thin-films on semiconductor substrates is increasingly
relevant. This dissertation has focused on new methods for applying self-assembly and
patterning techniques to germanium.
Chapter 2 discussed our development of a one-pot synthesis technique for
self-assembly on germanium. Previous studies explained multi-step processes for the
removal of germanium’s native oxide layer, passivation of its surface by chemical
etching, and solution deposition of alkanethiol monolayers. By controlling deposition
conditions, we were able to remove the oxide and deposit the SAM simultaneously, in a
single step, simplifying the process. We were able to show conclusively that oxide
removal, and not surface passivation, is the critical factor for self-assembly on
germanium. In Chapter 3, we described the ability to apply soft lithography techniques to
patterning germanium substrates. Incomplete monolayer coverage remains a challenge,
presumably a result of surface re-oxidation before the monolayer has time to form.
Improving these techniques, and applying these methods, will simplify
self-assembly on germanium such that it will become a technologically relevant
technique for controlling germanium interfacial chemistry.
60
4.2 Future Work
Self-assembly on germanium is in its earliest stages, with significant work
remaining to develop germanium’s full range of capabilities. First, the inhibiting factors
in SAM formation must be identified, the sources of which must be addressed. Next,
new techniques must be incorporated both to characterize and to improve upon
monolayer formation.
Re-oxidation of the surface has been proposed as an important factor that both
limits SAM formation, and is responsible for the degradation of formed monolayers.6
X-ray photoelectron spectroscopy analysis confirms the presence of oxygen on samples
containing SAMs. We suggest that defect sites in the SAM allow oxygen to interact with
the germanium surface, allowing re-oxidation to occur. It is difficult to produce an
entirely defect-free monolayer, so monolayers of germanium will always be susceptible
to degradation. We propose that instead of using the thiolate monolayer as a passivating
agent, it could instead be used to functionalize the surface for application of
high-performance dielectric films, which would, in turn, prevent formation of
germanium’s oxide layer. It has already been established that alkanethiols exhibit the
ability to form monolayers on germanium.6, 25, 87, 88
They are also capable of being
terminated with numerous functional groups.12
If patterning methods improve to
incorporate hybrid techniques, such as microcontact insertion printing, control over the
placement and type of the adsorbate layers could potentially aid in the stability of the
monolayer and its ability to accept thin films, subsequently making it device-ready. The
success of such a strategy is reliant upon the ability to produce full-coverage and stable
monolayers via numerous chemical patterning techniques, a large library of molecules for
61
self-assembly, and the ability to characterize and to further manipulate the electronic
properties of the resulting sample. To accomplish this, it will be necessary to tweak and
to test new combinations of adsorbates, solvents, crystalline planes, insertion methods,
and analysis equipment.
4.2.1 Chemicals, Solvents, and Crystalline Planes
The ultimate motivation for studying monolayer deposition on germanium is the
potential for a new, faster, generation of electronic devices. It is necessary to establish a
reproducible and cost-effective method for device fabrication, so simple, cheap, and
atmospherically insensitive procedures and devices are required. Continued tinkering
with variable substrate and adsorbate combinations may enable production of such
apparatus.
Variations in chemical adsorbates, solvents, and crystalline planes have exhibited
the ability to alter sample quality. Wang and co-workers showed that the stability of
germanium nanowires increased linearly with alkyl chain length.89
They attributed the
increase in stability to stronger inter-chain van der Waals interactions, and superior
density of molecular films, a common characteristic of longer carbon chains. Greater film
density affords better oxygen blocking ability. Depositing C12 monolayers on
germanium nanowires produced a sample capable of resisting re-oxidation for 24 hours
of ambient exposure.89
Alkyl chains longer than C12 were not studied. If further analyses
show results consistent with those of C2-C12, a new standard for surface passivation may
be developed. We completed preliminary work that suggests a decrease in solubility as a
function of increase in chain length, where C18 failed to dissolve in solutions adequate
62
for C12. Novel chemistries will be required to solve this and similar problems. It is well
known that the ability of alkanethiols to dissolve in solution is dependent upon the type
of solvent. As a result, methanol exhibits an enhanced ability, compared to ethanol and
t-butanol, to enable SAM formation on germanium at both high- and low-water
concentration mixtures. This leads us to conclude that any adsorbate may be deposited if
optimal solution conditions are met.
The germanium crystallographic orientation can have a large effect on SAM
quality. Ardalan and co-workers showed superior monolayer quality on Ge(100),
compared to that on Ge(111).6 Determining the structure of the monolayer on each face
will enable us to design better assemblies on germanium surfaces. Other variables have
not been ruled out as possible factors in film quality. Further investigation is required to
determine the optimal conditions for SAM formation.
4.2.2 Extending Chemical Patterning
Incomplete monolayer coverage remains a challenge, the source of which must be
identified and addressed. Ultimately, we would like to extend other patterning techniques
to germanium to enhance adsorbate placement and chemical functionality. We believe
that extending characterization of soft lithography techniques will lead directly to the
development of more precise and efficient methods for technological applications. Both
microcontact insertion printing and microdisplacement printing are powerful techniques
for their abilities to increase molecular placement selectivity, to limit feature diffusion,
and to prepare multi-functional surfaces composed of numerous exposed functionalities.
Extending techniques comparable to those applied for chemical patterning on gold will
63
produce results similar to our work in microcontact printing. Careful analysis and the
development of new chemistry techniques will have a significant impact on future soft
lithography of many materials.
4.2.3 Infrared Reflection Absorption Spectroscopy (IRRAS)
Infrared reflection absorption spectroscopy is one of the most efficient tools for
characterizing the bonding and chemical interactions of molecules on surfaces. It is an
important tool for evaluating interface properties, structures, monolayer stability, and
functional group identification. This spectroscopic technique introduces infrared radiation
to the substrate surface at a grazing angle, exciting vibrational modes of surface species,
and reflecting from the surface to a detector. Adsorbed species absorb energy from the
incident radiation, proportional to that of vibrational modes. Vibrations correspond to
stretching modes induced by slightly differing energies as shown in Figure 4.1. As in
XPS, the surface species exhibit unique vibrational frequencies, thus, detected radiation
can identify the surface components.
We have utilized IRRAS for characterizing self-assembled monolayers on gold.
Figure 4.2 shows a typical IR spectra for C12 SAMs on Au{111}. This proved to be an
excellent diagnostic for analyzing the kinetic behavior of self-assembly on Au{111}.
Unfortunately, germanium’s transparency to and reflection of infrared radiation precludes
its facile use with grazing angle IRRAS experiments. To study monolayer properties,
adsorbate-substrate binding mechanisms, and functional groups on alkyl chains, we must
alter our techniques for Ge SAM analyses.
64
Figure 4.1: Stretching and bending vibrational modes of methylene (–CH2–) groups.38
rocking, or bending in-plane
~720 cm-1
65
Figure 4.2: Infrared reflection absorption spectroscopy spectra of a C12 SAM on
Au{111}, showing higher energy stretching modes (A), and lower energy bending and
wagging modes (B).38
A
B
66
Infrared radiation at near-normal incidence to a highly reflective metallic surface
will result in a standing wave, resultant from the interaction between incident and
reflected wave. The electric-field produced from the standing wave will exhibit an
amplitude of zero at the surface, inhibiting interaction between the surface and any
adsorbed layers, making obtainment of surface molecule spectra impossible.107
Monolayers of an array of adsorbates, including thiols, CO, and H2SO4, on metallic
surfaces, including platinum, gold, and various semiconductors, to name a few, have been
analyzed by IRRAS equipment configured with an attenuated-total-reflectance (ATR)
configuration.108-112
Attenuated-total-reflectance surface-enhanced infrared reflection
absorption spectroscopy (ATR-SEIRRAS), offers strong signal sensitivity and negligible
interference from bulk materials.113, 114
Attenuated-total-reflectance SEIRRAS
measurements have exhibited the ability to produce strong absorption spectra for samples
closely related to germanium; monolayers on silicon.108, 109
In ATR applications, a
medium of high refractive index is optically contacted with a medium of low refractive
index (an adsorbing medium).115
The high refractive medium acts as an internal reflection
element (IRE) and is transparent to infrared radiation. Incident infrared radiation
impinges upon the IRE and passes into the absorbing medium-IRE interface with an
angle of incidence greater than the critical angle. In the absence of an absorbing medium,
the light will be totally reflected at the interface.115
In the presence of an absorbing
medium, some reflection intensity will be lost in the form of evanescent waves.
Figures 4.3 and 4.4 show a schematic of the ATR assembly and a depiction of the
evanescent wave, respectively. Evanescent wave formation is explained by the
Schrödinger wave function, which states that the electric and magnetic fields cannot be
67
discontinuous at a boundary. In our case, the absorbing medium is a thiol monolayer, and
the IRE is the germanium substrate. Upon reflection at the boundary, a loss in incident
beam energy, proportional to the vibrational properties of the interface, occurs in the
form of evanescent wave formation.115
As a result, the collection of absorption spectra
becomes possible. The number of reflections is dependent upon material properties and
experimental values, such as angle of incidence, and incident beam polarization.115
The
nature of ATR produces a strong IR signal representative of the surface species and not
the substrate material. Therefore, it is a powerful technique for observing monolayer
properties on germanium.
4.3 Conclusions and Final Thoughts
The electronics industry is making a rapid transition toward a new generation of
fast, and highly diverse devices. Fabrication of such components requires materials
exhibiting both superior electronic characteristics, and the capability to perform at the
nanoscale. Germanium has excellent intrinsic electrical properties, and control of its
interfacial chemistry by self-assembly is powerful. Furthermore, it lends itself to a
number of modern applications. Complications involving its oxide layer, resulting in
instability and deposition difficulties, are currently being studied.
We have begun to address the issues of self-assembly on germanium. This
dissertation demonstrated that self-assembly and μCP can be applied to the Ge(100)
surface. Several factors inhibit SAM quality and we have gained some insight into the
solutions to these problems. Further analyses will improve upon this work and assist in
68
Figure 4.3: Schematic of the attenuated total reflectance assembly. An absorbing
medium is placed in intimate contact with an infrared transparent crystal, the internal
reflection element. The angle of incidence is set such that it exceeds the critical angle,
ensuring total internal reflection. The crystal edges are cut as to allow entrance and
emission of incident beam radiation.116
69
Figure 4.4: Schematic depicting the formation of an evanescent wave as a result of
total internal reflection. Upon contact with the IRE boundary, an electric field, in the
form of an evanescent wave will develop, as described by Schrödinger’s wave
function. In the presence of an absorbing medium, the incident beam will experience a
reduction in intensity, corresponding to the medium’s absorption of the evanescent
wave, making absorption spectra collection possible.117
70
the development of industry-ready techniques for SAMs on technological surfaces.
Eventually, the investigation of germanium’s capacity to interact with technologically
important materials may yield the stable, electronically-advanced device, required for the
next generations of electronics.
71
Appendix 1
Non-Technical Abstract
A.1 Non-Technical Abstract
The focus on faster, smaller, and more diverse electronic devices has prompted
research into new materials. The search is on for semiconductor substrates with superior
electronic capabilities. Germanium has emerged as a semiconductor material capable of
producing the performance required for semiconductor devices, but has materials
properties that have limited its application. Methods for improving the stability and
performance of electronic materials have been developed. In order to manufacture
electronic components, both designed from new materials and capable of integration into
existing devices, established methods for material manipulation must be adapted. This
dissertation will address the issues connected to extending established techniques to new
materials, particularly germanium. It is our hope that studying both the ability and
limitations of these techniques will lead to the development of industrially relevant
procedures and devices.
Germanium’s surface is characterized by instability. The chemical make-up of
germanium’s surface inhibits its ability to interact with materials necessary for device
fabrication, thus limiting its capacity for integration. Self-assembled monolayers (SAMs)
are the result of molecular adsorption on solid surfaces. The organization and chemical
characteristics of SAMs allow them to form on, to stabilize, and to control the properties
of a number of substrates. Successful deposition of SAMs on germanium can correct its
intrinsic limitations. This dissertation investigates techniques capable of achieving SAM
deposition. Additionally, it will examine procedures exhibiting the capacity to enhance
72
control of electrical properties as well as the ability to incorporate a broad spectrum of
relevant materials. In a process akin to stamping ink designs on paper, chemical
patterning deposits adsorbates onto the germanium surface in a formation representing
that of a patterned soft polymer stamp. This paper will identify successful techniques for
improving the limitations of germanium, identify additional shortcomings, and propose
solutions to some of the present problems.
73
References
(1) Mantooth, B. A.; Weiss, P. S. Proceedings of the IEEE 2003, 91, 1785-1802.
(2) Mullen, T. J.; Srinivasan, C.; Hohman, J. N.; Gillmor, S. D.; Shuster, M. J.; Horn,
M. W.; Andrews, A. M.; Weiss, P. S. Applied Physics Letters 2007, 90, 063114.
(3) Smith, R. K.; Lewis, P. A.; Weiss, P. S. Progress in Surface Science 2004, 75,
1-68.
(4) Tirrell, M.; Kokkoli, E.; Biesalski, M. Surface Science 2002, 500, 61-83.
(5) Kasemo, B. Surface Science 2002, 500, 656-677.
(6) Ardalan, P.; Musgrave, C. B.; Bent, S. F. Langmuir 2009, 25, 2013-2025.
(7) Loscutoff, P. W.; Bent, S. F. Annual Review of Physical Chemistry 2006, 57,
467-495.
(8) Misra, D.; Garg, R.; Srinivasan, P.; Rahim, N.; Chowdury, N. A. Materials
Science in Semiconductor Processing 2006, 9, 741-748.
(9) Moore, G. E. Electronics 1965, 38, 114-117.
(10) Kittel, C. Introduction to Solid State Physics, 7th ed.; John Wiley & Sons, Inc.:
New York, 1996.
(11) Adamson, A. W.; Gast, A. P. Physical Chemistry at Surfaces, 6th ed.;
Wiley-Interscience: New York, 1997.
(12) Love, J. C.; Estroff, L. A.; Kriebel, J. K.; Nuzzo, R. G.; Whitesides, G. M.
Chemical Reviews. 2005, 105, 1103-1169.
(13) Kramer, S.; Fuierer, R. R.; Gorman, C. B. Chemical Reviews 2003, 103,
4367-4418.
(14) Allara, D. L.; Nuzzo, R. G. Langmuir 1985, 1, 52-66.
(15) Nuzzo, R. G.; Allara, D. L. Journal of the American Chemical Society 1983, 105,
4481-4483.
(16) Dubois, L. H.; Nuzzo, R. G. Annual Review of Physical Chemistry 1992, 43,
437-463.
(17) Poirier, G. E. Chemical Reviews 1997, 97, 1117-1127.
74
(18) Poirier, G. E.; Tarlov, M. J. Langmuir 1994, 10, 2853-2857.
(19) Ulman, A. Chem. Rev. 1996, 96, 1533-1554.
(20) Nuzzo, R. G.; Zegarski, B. R.; Dubois, L. H. Journal of the American Chemical
Society 1987, 109, 733-740.
(21) Ulman, A.; Eilers, J. E.; Tillman, N. Langmuir 1989, 5, 1147-1152.
(22) Stranick, S. J.; Atre, S. V.; Parikh, A. N. Nanotechnology 1996, 7, 438-442.
(23) Stranick, S. J.; Parikh, A. N.; Tao, Y. T. Journal of Physical Chemistry B 1994,
98, 7636-7646.
(24) Yan, L.; Huck, W. T. S.; Whitesides, G. M. Journal of Macromolecular
Science-Polymer Reviews 2004, C44, 175-206.
(25) Hohman, J. N.; Kim, M.; Lawrence, J. A.; McClanahan, P. D.; Weiss, P. S. In.
Prep. Journal of the American Chemical Society 2010.
(26) Bent, S. F. ACS Nano 2007, 1, 10-12.
(27) Bermudez, V. M. Langmuir 2003, 19, 6813-6819.
(28) Zerulla, D.; Chasse, T. J. Electronics Spectroscopy 2009, 172, 78-87.
(29) Wolpers, M.; Viefhaus, H.; Stratman, M. Applied Surface Science 1990, 45,
167-170.
(30) Cimatu, K.; Baldelli, S. Journal of Physical Chemistry C 2007, 111, 7137-7143.
(31) Love, J. C.; Wolfe, D. B.; Chabinyc, M. L.; Paul, K. E.; Whitesides, G. M.
Journal of the American Chemical Society 2002, 124, 1576-1577.
(32) Schlenoff, J. B.; Li, M.; Ly, H. J. Journal of the American Chemical Society 1995,
117, 12528-12536.
(33) Poirier, G. E.; Pylant, E. D. Science 1996, 272, 1145-1149.
(34) Laibinis, P. E.; Whitesides, G. M.; Allara, D. L. Journal of the American
Chemical Society 1991, 113, 7152-7167.
(35) Schreiber, F. Progress in Surface Science 2000, 65, 151-256.
(36) Nishi, N.; Hobara, D.; Yamamoto, M.; Kakiuchi, T. Journal of Chemical Physics
2003, 118, 1904-1911.
75
(37) Yang, G. H.; Liu, G. Y. Journal of Physical Chemistry B 2003, 107, 8746-9759.
(38) Saavedra, H. M. Ph.D. Thesis, The Pennsylvania State University, University
Park, PA, USA, 2010.
(39) Cassie, A. B. D.; Baxter, S. Transactions of the Faraday Society 1944, 40,
546-551.
(40) Li, D.; Neumann, A. W. In Applied Surface Thermodynamics; Spelt, J. K., Ed.;
Marcel Dekker, Inc.: New York, 1996; Vol. 63, pp 109-168.
(41) Kwok, D. Y.; Gietzelt, T.; Grundke, K.; Jacobasch, H. J.; Neumann, A. W.
Langmuir 1997, 13, 2880-2894.
(42) Gennes, P. G. d.; Brochard-Wyart, F.; Quere, D. Capillarity and Wetting
Phenomena: Drops, Bubbles, Pearls, Waves; Springer Verlag: New York, 2004.
(43) MacRitchie, F. Chemistry at Interfaces; Academic Press, Inc.: San Diego, 1990.
(44) Kolasinski, K. W. Surface Science: Foundations of Catalysis and Nanoscience;
John Wiley & Sons, Inc.: Wes Sussex, 2002.
(45) Gaydos, J.; Neumann, A. W. In Applied Surface Thermodynamics; Spelt, J. K.,
Ed.; Marcel Dekker, Inc.: New York, 1996; Vol. 63, pp 169-238.
(46) Wenzel, R. N. Industrial and Engineering Chemistry 1936, 28, 988-994.
(47) McClanahan, P. D. Honors, Pennsylvania State University, State College, 2009.
(48) Spelt, J. K.; Vargha-Butler, E. I. In Applied Surface Thermodynamics; Spelt, J. K.,
Ed.; Marcel Dekker, Inc.: New York, 1996; Vol. 63, pp 379-411.
(49) Lander, L. M.; Siewierski, L. M.; Brittain, W. J.; Vogler, E. A. Langmuir 1993, 9,
2237-2239.
(50) Srinivasan, C.; Mullen, T. J.; Hohman, J. N.; Anderson, M. E.; Dameron, A. A.;
Andrews, A. M.; Dickey, E. C.; Horn, M. W.; Weiss, P. S. ACS Nano 2007, 1,
191-201.
(51) Qin, D.; Xia, Y. N.; Whitesides, G. M. Nature Proocols 2010, 5, 491-502.
(52) Xia, N.; Thodeti, C. K.; Hunt, T. P.; Xu, Q. B.; Ho, M.; Whitesides, G. M.;
Westervelt, R.; Ingber, D. E. Faseb Journal 2008, 22, 1649-1659.
(53) Wang, Y.; Mirkin, C. A.; Park, S. J. ACS Nano 2009, 3, 1049-1056.
76
(54) Saavedra, H. M.; Mullen, T. J.; Zhang, P.; Dewey, D. C.; Claridge, S. A.; Weiss,
P. S. Reports on Progress in Physics 2010, 73, 1-40.
(55) Delamarche, E.; Schmid, H.; Bietsch, A.; Larsen, N. B.; Rothuizen, H.; Michel,
B.; Biebuyck, H. Journal of Physical Chemistry B 1998, 102, 3324-3334.
(56) Libioulle, L.; Bietsch, A.; Schmid, H.; Michel, B.; Delamarche, E. Langmuir
1999, 15, 300-304.
(57) Larsen, N. B.; Biebuyck, H.; Delamarche, E.; Michel, B. Journal of the American
Chemical Society 1997, 119, 3017-3026.
(58) Wilbur, J. L.; Kumar, A.; Kim, E.; Whitesides, G. M. Advanced Materials 1994,
6, 600-604.
(59) Dameron, A. A.; Hampton, J. R.; Smith, R. K.; Mullen, T. J.; Gillmor, S. D.;
Weiss, P. S. Nano Letters 2005, 5, 1834-1837.
(60) Moulder, J. F.; Stickle, W. F.; Sobol, P. E.; Bomben, K. D. Handbook of X-ray
Photoelectron Spectroscopy; Perkin-Elmer Corporation: Eden Prairie, 1992.
(61) Prabhakaran, K.; Maeda, F.; Watanabe, Y.; Ogino, T. Applied Physics Letters
2000, 76, 2244-2246.
(62) Rivillon, S.; Chabal, Y. J. Applied Physics Letters 2005, 87, 253101-253103.
(63) Prabhakaran, K.; Ogino, T. Surface Science 1995, 325, 263-271.
(64) Hohman, J. N.; Lawrence, J. A.; Weiss, P. S. Unpublished Results 2010.
(65) Han, S. M.; Ashurst, W. R.; Carraro, C.; Maboudian, R. Journal of the American
Chemical Society 2001, 123, 2422-2425.
(66) He, J.; Lu, Z.-H.; Mitchell, S. A.; Wayner, D. D. M. Journal of the American
Chemical Society 1998, 120, 2660-2661.
(67) Kosuri, M. R.; Cone, R.; Li, Q.; Han, S. M. Langmuir 2004, 20, 835-840.
(68) Cullen, G. W.; Amick, J. A.; Gerlich, D. Journal of the Electrochemical Society
1962, 109, 124-127.
(69) Lu, Z. H. Applied Physics Letters 1996, 68, 520-522.
(70) Roche, J.; Ryan, P.; Hughes, G. J. Applied Surface Science 2001, 174, 271-274.
77
(71) Gothelid, M.; LeLay, G.; Wigren, C.; Bjorkqvist, M.; Rad, M.; Karlsson, U. O.
Applied Surface Science 1997, 115, 87-95.
(72) Weser, T.; Bogen, A.; Konrad, B.; Schnell, R. D.; Schug, C. A.; Steinmann, W.
Physical Review B 1987, 35, 8184-8188.
(73) Kosuri, M. R.; Cone, R.; Li, Q.; Han, S. M. Langmuir 2004, 20, 835-840.
(74) Shirley, D. A. Physical Review B 1972, 5, 4709-4714.
(75) Bain, C. D.; Troughton, E. B.; Tao, Y. T.; Evall, J.; Whitesides, G. M.; Nuzzo, R.
G. Journal of the American Chemical Society 1989, 111, 321-335.
(76) Bain, C. D.; Whitesides, G. M. Science 1988, 240, 62-63.
(77) Nuzzo, R. G.; Dubois, L. H.; Allara, D. L. Journal of the American Chemical
Society 1990, 112, 558-569.
(78) Laibinis, P. E.; Bain, C. D.; Whitesides, G. M. Journal of Physical Chemistry
1991, 95, 7017-7021.
(79) Folkers, J. P.; Laibinis, P. E.; Whitesides, G. M. Langmuir 1992, 8, 1330-1341.
(80) Biebuyck, H. A.; Bian, C. D.; Whitesides, G. M. Langmuir 1994, 10, 1825-1831.
(81) Tabet, N.; Faiz, M.; Hamdan, N. M. Surface Science 2003, 523, 68-72.
(82) Hanrath, T.; Korgel, B. A. Journal of the American Chemical Society 2004, 126,
15466-15472.
(83) Porter, L. A.; Choi, H. C.; Ribbe, A. E.; Buriak, J. M. Nano Letters 2002, 2,
1067-1071.
(84) Wojtyk, J. T. C.; Tomietto, M.; Boukherroub, R.; Wayner, D. D. M. Journal of
the American Chemical Society 2001, 123, 1535-1536.
(85) Ivanisevic, A.; Mirkin, C. A. Journal of the American Chemical Society 2001,
123, 7887-7889.
(86) Maynor, B. W.; Filocamo, S. F.; Grinstaff, M. W.; Liu, J. Journal of the American
Chemical Society 2001, 124, 522-523.
(87) Elshocht, S. V.; Brijs, B.; Caymax, M.; Conard, T.; Chiarella, T.; Gendt, S. D.;
Jaeger, B. D.; Kubicek, S.; Meuris, M.; Onsia, B.; Richard, O.; Teerlinck, I.;
Steenbergen, J. V.; Zhao, C.; Heyns, M. Applied Physics Letters 2004, 85,
3824-3826.
78
(88) Wang, D.; Dai, H. Applied Physics A 2006, 85, 217-225.
(89) Wang, D.; Chang, Y. L.; Liu, Z.; Dai, H. Journal of the American Chemical
Society 2005, 127, 11871-11875.
(90) Stanley, S. K.; Joshi, S. V.; Banerjee, S. K.; Ekerdt, J. G. Journal of Vacuum
Science and Technology A 2006, 24, 78-83.
(91) King, T. J.; Saraswat, K. C. Journal of the Electrochemical Society 1994, 141,
2235-2241.
(92) Zanier, S.; Guldner, Y.; Berroir, J. M.; Vieren, J. P.; Faini, G.; Cambril, E.;
Garchery, L.; Campidelli, Y.; Sagnes, I. Thin Solid Films 1997, 294, 315-317.
(93) Okamoto, S.; Maruyama, E.; Terakawa, A.; Shinohara, W.; Nakano, S.;
Hishikawa, Y.; Wakisaka, K.; Kiyama, S. Solar Energy Materials & Solar Cells
2001, 66, 85-94.
(94) Izyumskaya, N.; Alivov, Y.; Morkoc, H. Critical Reviews in Solid State and
Materials Sciences 2009, 34, 89-179.
(95) Monget, C.; Schiltz, A.; Joubert, O.; Vallier, L.; Guillermet, M.; Tormen, B.
Journal of Vacuum Science and Technology B 1998, 16, 1833-1840.
(96) Rao, R.; Bradby, J. E.; Williams, J. S. Applied Physics Letters 2007, 91,
1231131-1231133.
(97) Tillack, B.; Richter, H. H.; Ritter, G.; Wolff, A.; Morgenstern, G.; Eggs, C.
Journal of Vacuum Science and Technology B 1996, 14, 102-105.
(98) Lopinski, G. P.; Eves, B. J.; Hul'ko, O.; Mark, C.; Patistas, S. N.; Boukherroub,
R.; Ward, T. R. Physical Review B 2005, 71, 1253081-1253084.
(99) Choi, K.; Buriak, J. M. Langmuir 2000, 16, 7737-7741.
(100) Kim, H.; McIntyre, P. C.; Chui, C. O.; Saraswat, C.; Cho, M. H. Applied Physics
Letters 2004, 85, 2902-2904.
(101) Magagnin, L.; Maboudian, R.; Carraro, C. Journal of Physical Chemistry B 2002,
106, 401-407.
(102) Mirkin, C. A. ACS Nano 2007, 1, 79-83.
(103) Su, M.; Liu, X.; Li, S. Y.; Dravid, V. P.; Mirkin, C. A. Journal of the American
Chemical Society 2002, 124, 1560-1561.
79
(104) Wang, D.; Thomas, S. G.; Wang, K. L.; Xia, Y.; Whitesides, G. M. Applied
Physics Letters 1997, 70, 1593-1595.
(105) Dameron, A. A.; Hampton, J. R.; Gillmor, S. D.; Hohman, J. N.; Weiss, P. S.
Journal of Vacuum Science and Technology B 2005, 23, 2929-2932.
(106) Bessueille, F.; Pla-Roca, M.; Mills, C. A.; Martinez, E.; Samitier, J.; Errachid, A.
Langmuir 2005, 21, 12060-12063.
(107) Greenler, R. G. Journal of Chemical Physics 1966, 44, 310-315.
(108) Hartstein, A.; Kirtley, J. R.; Tsang, J. C. Physical Review Letters 1980, 45,
201-204.
(109) Scheres, L.; Giesbers, M.; Zuilhof, H. Langmuir 2010, 26, 4790-4795.
(110) Nishikawa, Y.; Fujiwara, K.; Ataka, K.; Osawa, M. Analytical Chemistry 1993,
65, 556-562.
(111) Bjerke, A. E.; Griffiths, P. R.; Theiss, W. Analytical Chemistry 1999, 71,
1967-1974.
(112) Wandlowski, T.; Ataka, K.; Pronkin, S.; Diesing, D. Electrochimica Acta 2004,
49, 1233-1247.
(113) Roseler, A.; Korte, E. H. Fresenius Journal of Analytical Chemistry 1998, 362,
51-57.
(114) Huo, S. J.; Xue, X. K.; Yan, Y. G.; Li, Q. X.; Ma, M.; Cai, W. B.; Xu, Q. J.;
Osawa, M. Journal of Physical Chemistry B 2006, 110, 4162-4169.
(115) Ekgasit, S.; Ishida, H. In Handbook of Vibrational Spectroscopy; Griffiths, P. R.,
Ed.; John Wiley & Sons, Ltd.: New York, 2002; Vol. 2, pp 1508.
(116) Attenuated Total Reflectance (ATR); Pecsa Analytical, 1990.
(117) Martin-Gil, J.; Palacios-Leble, G.; Ramos, P. M.; Martin-Gil, F. J. Journal of
Interdisciplinary Celtic Studies 2007, 5, 63-76.